Lubrication Notes (upd 03/08)

Moderators: Amskeptic, Sluggo

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Lubrication Notes (upd 03/08)

Post by Amskeptic » Sat Feb 28, 2015 7:23 pm

There appear to be divergent opinions about the purpose and function of the oil pressure relief and oil control valves in our VW engines.

1. Oil Pressure Relief Valve - located at left rear of engine

2. Oil Control Valve - located at front left of Type 1 engine
(Type 4 is located between #1 lifter bores)


Lubrication of our engines depends on a supply of pressurized oil through galleries drilled in the case and the crankshaft to the main, rod, and cam bearings, and via the lifters to hollow pushrods to the rocker arm bushings and valve adjustment screws. Downstream components like cam lobes, gears, flywheel shims, pistons, cylinder walls, and distributor bushings, get their lubrication from spray coming off the main and connecting rod bearings.

1 ) Volume
The quantity of oil brought to a specific number of rotating journals/bearings required to maintain a hydrodynamic film and rinse away heat and debris, is met by the pump's output volume. It is calculated at the drawing board and established throughout the engine by the selected diameters of both the main gallery and local supply galleries and specified bearing clearances at the recommended oil viscosity. The VW engine uses a "positive displacement" pump that moves a specific volume of oil at every rotation. It squirts roughly 2 gallons per thousand rotations. At idle (450 camshaft rpm), that is just under 1 gallon per minute to distribute through the case galleries to the following
Pressurized Lubrication Points:
* 8 lifters > pushrods > 8 rocker arm bushings/shafts > all 8 valve adjusting screws
* 3 camshaft bearing journals
** 4 main bearing journals, the front three of which also feed:
* crankshaft galleries which supply
** 4 connecting rod bearing journals

Because positive displacement pumps can crank out serious pressure without a second thought, you WILL note that oil pumps are sized for VOLUME not pressure. Required volume is dictated by the number of bearing surfaces the engine has to lubricate.

Note: system pressure ensures the delivery of required volume to the bearing. Once oil has arrived at the bearing, the working film ( about .0002" thick - less than a tenth of your bearing clearance) is maintained by a "wedge" of oil developed by rotation. This film/wedge is not dependent on system pressure, but rather the volume of oil that made it in between the journal and bearing surfaces. Think: water-skiing. The lubrication system has to fill the swimming pool, yes, but the water skier has to move to ride up on the water. (p.s. a big slalom ski board requires less water. VWs have very big boards)

2 ) Pressure
System pressure ensures the delivery of the required volumes of oil to all of the bearings. In the perfect alternate universe, pressure would be the sum total of pump volume hitting the resistances at our bearing clearances. But, pressure is more complex than that. We have to calculate the flow resistance through the galleries, all the corners, the filter media on the Type 4s, the oil cooler adapter on the horizontal Type 3 and Type 1 doghouse offset coolers, etc.

System pressure is solely determined by the engineering downstream of the selected pump volume. Don't blame the pump for the lousy oil pressure you hate, you'll lose sight of the more likely reasons for insufficient lubrication. All pressure does for us is ensure adequate supply to the far reaches of the system. Correct pressure is merely having our pump volume equitably distributed to all of our bearings at the correct flow rates. Blaming the pump right off leads you into the marketing labyrinth of tho$e who are happy to $ell you more pump capacity when your issues are very much elsewhere. If you keep passing out when you stand up, don't select the heart transplant first.

The above asteriks point to potential operating pressure deterioration over time (wholly separate from acute failures like failed gallery plugs blown up coolers, stripped out strainer plate screws, and that ol' favorite, a missing fuel pump push rod/sleeves in early Type 4 cases).
Double-asteriks denote potential serious system pressure drop caused by sloppy clearances that drain the swimming pool away on our hapless no-longer-water but more like molten-metal skier ...

Note: this is a highly engineered balancing act! Gallery diameters assist in the division of specific volumes to each bearing's supply requirement.

[ we interrupt this program for the following Type 1 Main #4 Gallery Restriction bulletin:
Since the 1960 introduction of the four main bearing crankshaft, the restrictor in the #4 main bearing oil feed is critical to maintain correct flow rate (read: not too much) to the bearing. This is the only main bearing that does not have to feed a connecting rod journal through a crankshaft drilling. Enjoy better oil pressure AND happier bearings AND less spraying out from around your pulley . . . ]


Now lets confound our highly engineered balancing act with variable temperature and rpm and of course the one-way ticket to sloppy piddling old age, wear. Galleries that are "too small" when the engine is cold, new, or at high rpm, are "too big" when the engine is hot, old, or at low rpm.

The three variables we have to chase all over the place:

a) bearing clearances (from cold to hot/new to old)
b) pump speed > volume (from idle to redline)
c) viscosity changes (from cold to hot).


aa) oil flow "X" increases exponentially through the main, rod, and camshaft bearings as clearances open up when hot, magnified as the engine ages. IF:
.002" = X @
.003" = 5X @@@@@
.006" = 25X @@@@@@@@@@@@@@@@@@@@@@@@@

bb) if we have sufficient supply at idle and when hot, we will have too much at speed and when cold. Bearings do not consume oil in as linear a fashion as pump output from idle to redline. Loose bearing clearances just gobble volume and kill pressure at high rpm.


cc) our old air-cooled engines were carefully designed with matched pump size to gallery size to bearing surface area/clearance using specified oil viscosities. The relief and control valve spring pressures, relief and control piston surface area exposure to galleries, dump port diameters, were all engineered to provide acceptable service across the full range of operating conditions.


ILLUSTRATION AA - Type 1 Engines Through 1967
8mm galleries, 17/19 mm pump gears. Single relief valve with dump.
Image

Here is a cold engine with the relief valve opened. Please note that total system pressure does act upon the cooler at all times. Note too, that cold/cool oil does not/may not flow through the cooler because of little to no pressure *differential* between the "in" and the "out" when relief valve is open. It is given the choice to do so, however, so as the engine warms up, the cooler slowly develops flow. When you expose the inlet of the cooler exclusively to the pump, and the outlet to bearing leak down, as when the relief valve closes, you will then have exclusive circulation through the cooler. According to factory specified volume/viscosity/resultant pressures, this switching occurs at a very nice 180* in the 48ish psi range.

This lubrication system just barely spanned the above listed variables but it worked acceptably for toodling around with 25/36/40 hp. As with all air-cooled engines that use alloy crankcases, hot or worn engines idle with low oil pressure. The substantial bearing surface area does not mind a barely 2 psi oil pressure . . . and VW owner's manuals state, "don't worry about a little flickering of the oil light at idle". So don't.



ILLUSTRATION B - Grooved Relief Piston 19/21 mm pump gears
Image


The introduction of the Type 3 1500S engine, the little 66 hp hotrod of the line-up at the time, arrived with a grooved oil relief piston. The physics of the grooved relief valve piston never was explained as to how it improves oil cooling. Subsequently, our internet universe has promulgated any number of sometimes cursory if not creative explanations. How does it actually work? How does it actually help oil cooling?
After all, the groove clearly only *works* when the valve is closed ...

This improvement gave us the first utilization of "waste circulation" whereupon the cooler is cooling off a greater quantity of oil than the bearings are consuming. When fully closed, the relief valve groove opens the main gallery to the dump port, delivering oil cooler-cooled unused oil back to the sump.

Extremely Exaggerated Example:
Old Days > Bearings consume four quarts per minute. That's four quarts sent through the cooler, then through the bearings, then those very same four quarts becomes hot oil dumping into the sump to then get picked up and sent through the cooler and back to the bearings again.

New Days > Bearings consume four quarts per minute. Grooved relief valve leaks a quart per minute. That is five quarts now going through the cooler for every four heated through the bearings. That would be a whole gallon of extra cooling every four minutes compared to the old system.
Great idea! The bad news is, this waste circulation doesn't know how to shut off in a hot engine with diminished oil pressure. Same problem as the prior designs, hot and older engines will flicker their idiot lights, the idiots.

The next improvement was a 19 mm oil pump that came along in 1967 with the new forged cross-drilled crankshaft that fed more oil to the main and connecting rod bearings. Very incremential additional flow was proferred as we owners flogged those little horses a little harder and a little faster.



ILLUSTRATION C - Type 1 Dual Relief Valves And 24mm Super Pump
Image


Introduced in the 1970 model year with a dished-in camshaft gear so the pump could fit, this design improvement expanded upon that trick "waste circulation" in a whole new way. With a super pump and a newly dedicated "control" dump valve at the other end of the main gallery, VW got rid of the grooved relief valve piston to allow total switch-off of waste circulation when the pressure dropped.
(note that the Muir book and others erred when they did not list that the control valve was introduced simultaneously with the super pump. Indeed, a 26mm pump was introduced the very next year in 1971, but that was a "very incremential" bump-up to help maintain FLOW through the new doghouse oil cooler adapter)

Now we have cold engines that can:
a) get rid of excess cold oil volume-thus-pressure,
b) allow even more waste oil circulation through the cooler than the engine actually consumes, and the control valve can close off the waste circulation! yay! at the first sign of hot/worn engine pressure drop when the bearings expand due to heat/age.
This illustration shows a fresh warm engine with its "about 16.0 pounds of spring pressure" closed relief valve (so the oil cooler can cool all of the oil), but at the same time, the "about 8.5 pounds of spring pressure" control valve is bleeding off the excess capacity oil volume from the 24mm pump because it still has nice .002" bearing clearances.

Similar to HVAC heating and cooling thermostats, this new control valve has a "cut-in and "cut-out" differential to prevent too much cycling. This valve works in tandem with the relief valve to bias flow through the cooler. Get your spring pressures right!

As per jimmy111's excellent February 17, 2008 write-up on theSamba,

Dual Relief Oiling System. How it works.
http://www.thesamba.com/vw/forum/viewto ... 80ae8b1ffc

the small closed-valve surface area exposed to the gallery means that it opens at about 50-60 psi, a good pressure to open at, but once pushed down, the full surface area of the piston is exposed to gallery pressure . . . it will not close again until the pressure drops to about 28 psi. This roughly "brackets" the system operating pressure specifications of 28-42 psi. It also insures that dedicated oil cooler flow is maintained with a warm engine, because the relief valve needs to close ahead of the control valve. If you want to screw with the spring pressure or botch the seat profiles here, be prepared for the *relief* valve to wear prematurely in its bore as it becomes the oil equivalent of the little red UniSyn carb airflow indicator forever bobbing in its bore to the vagaries of engine rpm.


ILLUSTRATION D
But What About? . . . ex. 1
Image

But what about a leaky pressure relief piston? Not a big deal, say I. With a healthy engine, there is plenty of excess capacity to allow gallery oil to leak between the wall of the piston and the bore. It will act somewhat like a "grooved 1500 pressure relief piston." That is all.


ILLUSTRATION E
But What About If It Is Worse Than That? ... ex. 2
Image

But what about a leaky pressure relief piston in an older engine? See all of those oil drips cascading out from the bearings in the illustration? They reflect what was written further up. Worn bearings (at the factory wear limit) consume up to *25 times the volume* that fresh bearings do. Oil leaking from the main gallery to the dump port at the relief valve does not change the fact it still opens when pressure is (thankfully) high enough, and it still closes when pressure drops. However much the system pressure may further drop cannot be addressed by a stronger spring pressing against an already closed! relief piston dribbling oil from the gallery to the dump port and vent hole. The quantity of oil that is able to flow past the piston wall and the relief bore wall is substantially less than the leak of worn bearings.

(p.s. just did an experiment on my 1970 dual relief engine -
a) no spring in relief bore yielded no change in oil pressure, as piston at the bottom of the bore blocked the vent hole
b) removed piston entirely and had no oil pressure at idle, light would go out at 2,000 rpm )


ILLUSTRATION F
Type 4 Engine Hey It Has A 24mm Pump Operating Temperature
Image

The Type 4 engine re-utilizes the grooved relief valve plunger bleeding off excess volume and has a control valve located along the right lifter gallery between #1 lifter bores and it has massive bearing surface area . . . and a modest 24mm pump, even with the new oil filter. Did I mention that increased bearing area reduces the need for bucketloads of oil? I just did. But what about the 85 hp Porsche 914 shifting at 5,799 rpm? Yeah, same difference.
This illustration is of a typical happy Type 4 engine running down the road with "about 17 pounds spring pressure" having already closed the relief valve to make sure that the oil cooler has a dedicated oil flow from pump to the bearings and the "about 4 pounds spring pressure" control valve dump is still open, and the grooved-yet-again relief piston is bypassing some oil to the relief bore's dump.

[ we once again interrupt this broadcast to announce the following Type 4 Oil Filter Bypass bulletin:
There is another relief valve inside the oil filter mount that opens at only 9-12 psi.This does not mean that it opens every time there is more than 9 psi oil pressure in the engine. A well-maintained filter can flow 4 gpm of factory specified viscosity at 34*, that is what your positive displacement pump flows at 3,500 rpm dead cold, so don't . . . do that. Only if/when the filter cannot maintain a *flow rate* that is within a 9-12 psi differential between the filter's own "in" and "out" will the relief open. This occurs only very briefly when the engine is cold and the galleries haven't yet blubked up to full cold pressure, and during hot low viscosity high speed operation when you haven't changed your filter since forever and you deserve to die anyway. As George Nehls, P.E. wrote of a conversation between Gary Cross of Melling Automotive who communicated in a phone call with Gary Bilski of Allied Signal who punched out the numbers on his calculator, "something between one ten thousandth and one one hundred thousandth of the total oil flow in sixty hours of operation bypasses the oil filter." Type 1 engines have been running around for over sixty years without even possessing an oil filter , so, like, don't worry about it? ]



ILLUSTRATION G
Older Type 4 Engine
Image

A warm older engine is depicted here by the cascade of little drips pouring off the clattery bearings. The control valve is hardly ever asked to bypass any more in the warm engine now that it is in its dotage. There is some bypassing going on at the relief valve main gallery port-to-dump groove, but lower system pressure means that the relief spring can easily keep the piston closed so all the oil is at least run through the cooler for "viscosity enhancement".

So here is a question:
What on Earth does a heavier spring at the relief or control valve do for low operating oil pressure?

Answer: Nothing.

With factory correct spring tensions, those valves are shut below let's say 28 psi . . . and they will not push open until 60 psi where your engine may not be able to go anymore anyway.
Do you think that heavy springs make them more shut? Really really shut? If you have low oil pressure, *the springs are out of the loop*.


ILLUSTRATION H
Tired Old Type 4
Image

Here is a Type 4 engine with loose bearing clearances, loose relief piston, and even a loose control piston. All components are exactly where they belong, pistons are seated, slop between piston walls and bores only allow some oil flow and the vent holes only allow so much to pass at the bottom of the bores. The relief valve dump is doing pretty much what it has been doing since the engine was new, the control valve dump only passes what can escape between the piston wall and the bore. The primary cause of low oil pressure at idle is the leak rate past the bearings. When you read "bearings", remember that we are talking about 11 rapidly spinning journals, 8 lifters, 8 rocker arm bushings, all capable of passing 25 times the volume they used to.

[we interrupt this program for the following Type 4 Control Valve bulletin . . .
Type 4 control valves were NOT used in hydraulic lifter engines for the simple reason that the lifters happily utilized cold oil pressure to pump themselves up if and as necessary. Hydraulic lifters are continually bleeding themselves in operation at every down ramp off the cam lobes. I speculate that if you revert to solid lifters in a hydraulic non-control valve crankcase, the relief valve might spend too much time towards the bottom of its bore, playing "high rpm control valve", diverting oil that should be making a dedicated loop through the oil cooler ]



ILLUSTRATION I
A Modified Relief Valve In What Would Be The Warm Operating Or Shut Off Position But . . .
Image

On February 14, 2015, a customer and I had a substantial oil leak located just above the relief valve bore, seconds after starting a rebuilt 2.0 engine for the first time. The leak was coming out from a threaded steel gallery plug that looked to cap the dump bore drilling. As we were trying to figure out this leak, it spontaneously ceased as the engine warmed up to the 5- 10 minute point. That suggested having a look at the relief valve. I loosened the big slotted screw, keeping it pressed against the threads (you know what 19 lbs of relief valve spring pressure can do to soft aluminum threads on their last tur. . . never mind. There was *no* spring tension. I would have to assume that this is an error. A truncated grindedidy-offed half spring slid out. Then a BALL fell out.



ILLUSTRATION J
Modified Relief Valve Under Cold/High RPM
Image

I cannot figure this one out. I see a substantial bypass capability when the ball is pushed down, which is helpful but I wasn't able to see how the ball could "switch" the pump path to the main gallery when cold.


DISCUSSION
Because the air-cooled VW engine oil temperature and pressure control system utilizes viscosity and pressure in a precisely calibrated manner, you can be very sure that things change in a hurry if you start goofing with the hard-engineered pump capacity, spring tension/piston seating in relief or control valve, or oil viscosity.

Pick one of the below threshholds.
Your relief valve can't tell the difference between:

a) big 30mm pump = 220* oil due to late switchover to cooler
(note that at this instant, the oil is exactly the same pressure as below, but is hotter)
b) factory pump = 180* oil due to on-time switchover

c) monster relief spring = early switchover to cooler, but no difference to old or hot engine oil pressure
d) low viscosity 5w30 oil= early switchover to cooler and lower old or hot engine oil pressure

e)monster control spring > risk of cold oil cooler/gallery plug damage, excessive and unwanted relief valve action, not a lick of help to old or hot engine idle pressure



If you have low oil pressure at the specified rpm and viscosity, you have an imbalance between pump output and bearing leak-out. Many online vendors suggest installing a higher capacity pump or "heavy duty relief springs", primarily because they are selling higher capacity pumps and heavy duty relief springs. Heavier relief springs can blow up coolers by preventing the dump port from limiting cold system pressure, but you are still in your world of hurt at high temperatures.
High capacity pumps can just totally overwhelm the cooler and dump ports of both relief and control valves at cold temperatures, and they might serve as temporary bandaids while your engine suffers, I bring you that poor Vanagon 2.1 with the ovalized connecting rod that tried to warn the owner with diminishing oil pressure at highway speed. The new pump kept that warning buzzer off even when the #4 rod went through the case.


The most overlooked critical component to a healthy lubrication system is correct bearing clearances.

How many of us are building our engines with bearing clearances at the low end of the specification window? Are we instead just hitting somewhere in the range?

If factory main bearing clearance specifications call for .002" to .005" (wear limit .006"),
how many of our engines are coming in at .004"- .005", "good enough! It says SO right THERE" ?
If the factory is looking for .0008" to .0027" at the connecting rods, how many of us are landing at .003", "aw heck get off my back, this ain't no Raby engine . . ."

Are we nailing our camshaft clearances with these bearings from Brazil? Our engines require an eye-opening attention-getting .0008" to .002" (wear limit .0047") clearance, I'd aim for .001" to .0015" and make sure I washed my hands too.
Type 4 carbed engine cases that no longer have a fuel pump installed, need to be checked for a gaping fuel pump pushrod hole that intersects the right lifter gallery. If it has not been sleeved, well, you will be amazed at how much oil pressure you can not-have in an otherwise tight engine.

Some of us are 1/2 to 3/4 of the way through the expected service life of our fresh rebuild before it even leaves the bench. Look at those (shocking, I tell ya) oil consumption rates again. Right there, under the beautiful paint and new exhaust system and anticipatory hopefulness, we might have an engine that is relegated to maybe 50,000 miles saddled with questions about low oil pressure, while someone else gets 135,000 miles with an idiot light that has never come on at brutally hot temperatures AND a low idle.

I personally aim for .002" or .0025" on main bearings (#2 main slightly tighter, look at the Bentley), I shoot for a maximum end play of .003" at first start.
The connecting rods, well, .001" would be lovely, but because I have yet to work with a machinist who can vouch for the accuracy required for the factory-blessed .0008", I am a scairdycat .002". I did reassemble a Raby kit that had every connecting rods at a deadnuts accurate .0008" (unbelievable!), but it was unfortunately assembled by someone without proper regard for the very needed exquisite cleanliness, and those poor rods were initially bound to the crankshaft by contamination. When my customer and I were done, they slowly slowly glided down like a hydraulic strut-equipped Lexus glovebox door.

(p.s. tight bearing clearances do create extra heat, but the heat graph shows that you're cool enough at .0015" so while you are breaking in your .002" engine, just take it easy for 1,000 miles, you can spin er up briefly and frequently after the first 100 miles of course, but don't cruise at 4,000 rpm until everything is polished and friction-free)

That is the crux of it. Building an engine with serious loyalty to the exacting German engineer's specificity will erase many of the issues we contend with and then chase down numerous dead-end alleys. It demands that we "blueprint" and dot every "i" and cross every "t". If your bearing clearances check in sucky sloppy during your build up, bite the bullet and turn down the journals to the next undersize, but be prepared to pay attention through the entire process. You need your new undersize bearings beautifully and cleanly installed and torqued down, then measured accurately, then you have to somehow convince your machinist to come in at the high end of the specified undersize diameter (low end of the clearance) and polish to the best-accepted microfinish for your bearings with the "fuzz" opposite to the operating rotation (10 rs?), it goes on and on, but it is fun. Is there anyone left who will go the last mile with you?


Look at how that old 1971 vintage 26mm oil pump was stuck on every Type 1/3 engine and watercooled Vanagon from the factory up to 1991, the 24mm Type 4 pump handled Porsche 914s to 5,800 rpm, these pumps worked fine for decades within factory tolerances right through to the end of the engines' service lives, and you bet they did communicate through low pressure that it was time for a rebuild, but who was listening? We just kept driving, "what's that noise?", didn't we? At the next rebuild, the "only available" camshaft bearings had .004" clearance out of the box, the undersized connecting rod journals had to be cut a little extra low to clean up that deep groove, you know? and the #1 main bearing was loose in the crankcase bore, the new!-cuz-you-NEED-it! Shadek Leak-O-Matic oil pump washed out the Permatex between the case and the pump body, and sure, oil pressure becomes a problem when every galled rocker shaft has a bobbling rocker arm, but the rebuilder needed to get the work going out the door, "did I put a new o ring on that oil pick-up?" "oh, don't worry about that oil light at idle, the engine is just running a little hot during break-in."

I have been unbelievably blessed with a lifetime of no lubrication problems with my stock engines in 30 years, I did blow a gallery plug at 417,000 miles on the Road Warrior, but that was for other reasons)
I like the factory lubrication systems. They have never let me down.
Colin

kreemoweet
Getting Hooked!
Location: Seattle
Status: Offline

Re: Lubrication Notes

Post by kreemoweet » Sun Mar 01, 2015 1:23 pm

What a welcome relief this treatise is, from the usual morass of misinformation, speculation, and confusion that this topic generates elsewhere! I have a few remarks/questions:

1. I do not think that the Type I dual-relief cases have a "dump port" in the relief valve bore - at least there haven't been in any in the cases I've looked at. This makes sense, as the pressure control valve at the other end of the case took over it's function. At the same time, the use of the grooved relief plunger (311-115-411) was discontinued. Since VW went back to using the same grooved relief plunger in the '72 on Type 4 bus cases (as my VW Type 2 parts book attests), I assume the dump port was reinstated also.

2. Is it actually the case that the grooved relief plunger bypasses oil in the completely closed position as indicated? I have neither single-relief Type I, or Type 4 cases available to measure, but on my dual-relief Type I case, the bottom of the groove just barely intercepts the top of the main oil gallery port in that position, and I somehow have the notion that the bypass port was located somewhat lower than that. If it were up to me, I certainly would try to ensure there was a certain minimum oil pressure
available before I allowed oil to start spilling off thru the bypass port!

3. Does anyone know the original use (or Vw part #) of the grooved plunger with the sloped lower surface (as shown here: http://www.texasaircooled.com/catalog/i ... 115156.jpg? This plunger is being widely sold for use in Type I dual-relief cases (quite inappropriately, needless to say) by many VW parts vendors. One I have at hand is
23.75 mm long. Can anyone verify if this is the same length as the original straight-grooved plunger?

4. Regarding the topic on thesamba.com started by "Jimmy111" referred to: I do not find much at all believable in that post. In particular, I think the analyses given on
pressures acting on the relief plungers to be extremely dubious. I think it a mistake to regard the tops of the relief/control plungers and their seats in the case to be any
kind of sealing surface. Those surfaces on the (stock) parts I have looked at are rough-machined, not at all what one would expect at a "seal". Instead of the rather bizarre
discontinuities of force advocated there, I believe one should simply consider the (local) oil pressure to be operating uniformly on the entire top surfaces of the plungers.

5. It is somewhat difficult to reconcile the view of these lube systems as being "precisely calibrated" with the surprisingly large spec. tolerances given by VW. For instance,
the relief plunger spring pressure specs (5.5-9.2 lb for single-relief Type I engines, 12.3-16 lb for dual-relief Type I engines). Nevertheless, they seem to be precise enough
for the purpose.

User avatar
airkooledchris
IAC Addict!
Location: Eureka, California
Contact:
Status: Offline

Re: Lubrication Notes

Post by airkooledchris » Mon Mar 02, 2015 11:35 am

just want to receive updates to this thread going forward.

Thank you for putting this together.
1979 California Transporter

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: Lubrication Notes

Post by Amskeptic » Mon Mar 02, 2015 2:23 pm

kreemoweet wrote:What a welcome relief this treatise is, from the usual morass of misinformation, speculation, and confusion that this topic generates elsewhere! I have a few remarks/questions:

1. I do not think that the Type I dual-relief cases have a "dump port" in the relief valve bore - at least there haven't been in any in the cases I've looked at. This makes sense, as the pressure control valve at the other end of the case took over it's function.
There may be design differences . . . these are three AS 41 photographs off the internet. I believe the lowest hole is the piston's "anti-hydrostatic lock" vent:

Image

kreemoweet wrote:
At the same time, the use of the grooved relief plunger (311-115-411) was discontinued. Since VW went back to using the same grooved relief plunger in the '72 on Type 4 bus cases (as my VW Type 2 parts book attests), I assume the dump port was reinstated also.
The existence of the dump port was not in lock-step with grooved pistons, however. The relief valve dump port was the sole over-pressure release from 1930 on up to 1970 in Type 1 engines.

The Type 4 dump was a magnificent mess with two slotted longitudinal release areas down to the retaining screw. Late Type 4 cases had a whole dump *chamber* leading over to the oil pump inlet right case half when they introduced the new shorter oil pump pick up.
kreemoweet wrote: 2. Is it actually the case that the grooved relief plunger bypasses oil in the completely closed position as indicated? I have neither single-relief Type I, or Type 4 cases available to measure, but on my dual-relief Type I case, the bottom of the groove just barely intercepts the top of the main oil gallery port in that position, and I somehow have the notion that the bypass port was located somewhat lower than that. If it were up to me, I certainly would try to ensure there was a certain minimum oil pressure
available before I allowed oil to start spilling off thru the bypass port!
A grooved relief piston was not offered in the dual relief Type 1 application. The Type 3 engine was single relief when grooved piston was introduced. It apparently was effective enough that VW suggested not using it in retrofit applications in cold climates.

When you hope "to ensure there was a certain minimum oil pressure available before I allowed oil to start spilling off thru the bypass port!" I assure you that VW ensured that there was *sufficient oil pump capacity* at *maximum allowable bearing clearances* within *the mandated viscosity range*, and you do not, I repeat, you do not have to worry about minimum oil pressure. Especially with the 24mm gears, VW created a surplus supply. We all have been clouding up the physics with our innumerable pressure confusions and woes that are due to our own errors. What happens to the reputation of a beautiful lubrication system design when Present Day Rebuilder uses Sloppo'Fit Shadek and starts spewing about how VW made an inadequate system in the first place? Marketers LOVE that, "you NEED our pump!"
kreemoweet wrote: 3. Does anyone know the original use (or VW part #) of the grooved plunger with the sloped lower surface (as shown here: http://www.texasaircooled.com/catalog/i ... 115156.jpg? This plunger is being widely sold for use in Type I dual-relief cases (quite inappropriately, needless to say) by many VW parts vendors. One I have at hand is
23.75 mm long. Can anyone verify if this is the same length as the original straight-grooved plunger?
Length of the piston is not the critical dimension. Distance from crown of piston to groove, and height of groove is. A longer piston has better support in the bore. The spring seat depth should be consistent among different pistons, but if it was different, it would affect tension. You know that VW had oversize pistons for relief bore repair? I read it somewhere . . . :silent:

You need to straighten me out in a hurry as to whether or not that bevel-grooved piston showed at all in a dual relief case before the hydraulic lifter Mexican Beetle. We both know that the bevel allows more waste circulation (which is FINE in a healthy system!)
kreemoweet wrote: 4. Regarding the topic on thesamba.com started by "Jimmy111" referred to: I do not find much at all believable in that post. In particular, I think the analyses given on pressures acting on the relief plungers to be extremely dubious. I think it a mistake to regard the tops of the relief/control plungers and their seats in the case to be any kind of sealing surface. Those surfaces on the (stock) parts I have looked at are rough-machined, not at all what one would expect at a "seal".
Surface area vs pressure of the control and relief valves, like wheel cylinder and master cylinder pressure analysis, stands on mathematics.
The numbers *observed and written down* are numbers that were observed and written down. I see no physics defying observations, but I am certainly open to another set of eyes.

The upper end doesn't have to "seal". The VOLUME of any leak past the top of the "VW" stamped piston surface has to travel past a broad seat up there at the top of the relief bore, and it is negligible, insignificant, not important, minor, and utterly irrelevant since the two optional paths go to the same bearing destinations is under the same pressure! When the valve is closed against the seat, it is just making the oil flow through the cooler, any leak to the main gallery doesn't change the system pressure. Any leak to the dump port is far far more likely to be coming from the main gallery/piston wall slop in the non-grooved application, note please, that the grooved relief piston is designed to leak.

Have you ever had to work through an automatic transmission control valve assembly? That is hydraulics versus pressures versus switchpoints versus viscosity in a dizzyingly beautiful dance of applied physics with ball bearing switched galleries and buckshot ball stops (and one way fluid diode balls!). The VW relief/control physics is just totally simple and it works, and I guess what motivated me to jump in is because I see as similar and unfortunate of a mangling of the original engineering here as I do with distributors, cooling systems, carburetion, and exhaust, because people love to claim deficient factory engineering when it is their own errors and omissions that create problems. (I would have replaced an oil pump before I discovered the missing fuel pump push rod in that Type 4 engine back in 2005)
kreemoweet wrote: Instead of the rather bizarre discontinuities of force advocated there, I believe one should simply consider the (local) oil pressure to be operating uniformly on the entire top surfaces of the plungers.
From the Crosby Engineering Handbook:
A typical configuration is shown in Figure F2-3. As the disc begins to lift, fluid enters the control chamber exposing a larger area of the disc (Figure F2-2) to system pressure. Because of the larger disc area exposed to system pressure after the valve achieves lift, the valve will not close until system pressure has been reduced to some level below the set pressure. The design of the control chamber determines where the closing point will occur
kreemoweet wrote: 5. It is somewhat difficult to reconcile the view of these lube systems as being "precisely calibrated" with the surprisingly large spec. tolerances given by VW. For instance,
the relief plunger spring pressure specs (5.5-9.2 lb for single-relief Type I engines, 12.3-16 lb for dual-relief Type I engines). Nevertheless, they seem to be precise enough
for the purpose.
The precise calibration I refer to is more along the lines of "hard engineering". , Gallery length, corners, diameters, pump volume requirements divided into expected bearing surface areas X expected and allowed clearances for temperature variations - viscosity effects . . . by the time you are fine-tuning system pressures, the window of acceptable is broad. As mentioned at the bottom of the article, what the hell is with the vast acceptable bearing clearances? Is it .0008" or is it .005"? VW covered as wide of a net of manufacturing tolerances for as great of a number of units for sale in as broad of an environmental envelope as possible within the confines of "cheap economy car". I think they did a masterful job, and I think with time and love on our hands, we ought to be able to blueprint if not just hit the window they specified. I will tell you that I have been increasingly disturbed by the shoddiness I am running across in so many fresh! rebuilt! VW engines.

I am embarrassed (but-not-really) to not ever have needed one single aftermarket "improvement" over the past thirty five years and damn close to one million miles of driving along the edge of a destruction that the glossy magazine purveyors allege their parts will protect me from.
Alas, in my observations and experiences with good people who love their Volkswagens over the past thirteen years, it is their "improvements" which are hastening the destruction of these engines.
2019 Annual Funding Drive!
https://itinerant-air-cooled.com/viewto ... 41&t=13755

BobD - 1978 Bus . . . . . . . . . . .111,145 miles
Chloe - 1970 bus . . . . . . . . . . . 206,812 miles
Naranja - 1977 Westfalia . . . . . 94,655 miles
Pluck - 1973 Squareback . . . . . 55,510 miles
Alexus - 1991 Lexus LS400 . . . 76,560 miles

Boxcar
Addicted!
Location: Itinerant Fledgling
Status: Offline

Re: Lubrication Notes

Post by Boxcar » Mon Mar 02, 2015 7:21 pm

Them theres a treatise.
Just sayin.
1975 003 Auto Westy L90D

repair!!!!aug2015
Jan/16 Bumped mixture a few notches richer. finally developing HP.


1.8L/LJet/Pertron DVDA+PertronixCompufire 42/36Ham Heads/AA 93mm pistons/barrels.Porsc.Swiv.Adjusters/CromoSteel pushrds/ Web 9550Cam/55cc chmbr.,035 squish,8.6:1CR/German Supply VWCanadaReman Rods/Schadek 26mmPump/vdo dualOP8/10#low sender/Quart Deep Sump
Backdate Htr bxs,reflanged 914 4into1. Two and three eighths inch collector,magniflow*muffler

User avatar
asiab3
IAC Addict!
Location: San Diego, CA
Contact:
Status: Offline

Re: Lubrication Notes

Post by asiab3 » Tue Mar 03, 2015 12:48 am

amskeptic wrote:You need to straighten me out in a hurry as to whether or not that bevel-grooved piston showed at all in a dual relief case before the hydraulic lifter Mexican Beetle. We both know that the bevel allows more waste circulation (which is FINE in a healthy system!)
My old engine was a UO (1970 Type 3 1600) duel relief case and it had the grooved relief piston linked above. However, the groove in the piston does not line up with any dump ports in the old case, or the AS41 Brazilian case I'm building on currently. The groove lines up only with the main bearing gallery and plug, (picture below.)

I hereby wildly speculate (based on my two engines here) that the dual relief cases can use the tapered-grooved relief piston. I do not think the bevel/groove actually allows for more waste circulation in a dual relief T1 case. Could it be VW trying to get their money's worth out of the tooling for the grooved pistons?

This would mean the single relief case dump ports would have to be lined up with the spring extended in a low-pressure scenario. Does anybody have one apart right now?

Robbie

Image
1969 bus, "Buddy."
113k miles with me.
292k miles on Earth.

User avatar
Bleyseng
IAC Addict!
Location: Seattle again
Contact:
Status: Offline

Re: Lubrication Notes

Post by Bleyseng » Tue Mar 03, 2015 7:54 am

It's been standard practice to replace with a piece of welding rod the spring on the second relief valve on all W, EA, EC, GA type 4's in 914's. The later GC cases don't have the second valve and are preferred cases as they have better oiling at higher rpms. At higher rpms combined with cornering a 914 can get oil starved as the oil has a hard time getting back to the sump from the heads and past the windage tray. The workarounds are to enlarge the holes in the windage tray, a tuna can and block the second oil valve.

The type 4 case change to delete the second valve was for providing more oil for the hydro liters.

Good article!
Geoff
77 Sage Green Westy- CS 2.0L-160,000 miles
70 Ghia vert, black, stock 1600SP,- 139,000 miles,
76 914 2.1L-Nepal Orange- 160,000+ miles
http://bleysengaway.blogspot.com/

Jivermo
IAC Addict!
Status: Offline

Re: Lubrication Notes

Post by Jivermo » Tue Mar 03, 2015 9:14 am

Pouring gasoline into this conflagration...

Image
Type 4-looking up the bore of the oil pressure relief valve-modified version.

Image
Modified cut spring and ball.

Image
Stock VW spring and piston next to cut spring and ball modification.

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: Lubrication Notes

Post by Amskeptic » Tue Mar 03, 2015 3:05 pm

asiab3 wrote: My old engine was a UO (1970 Type 3 1600) duel relief case and it had the grooved relief piston linked above. However, the groove in the piston does not line up with any dump ports in the old case, or the AS41 Brazilian case I'm building on currently. The groove lines up only with the main bearing gallery and plug, (picture below.)
Robbie, a "duel" dual relief crankcase is just a dual relief crankcase.

Is it possible that it was a retrofitted relief valve in your old 1600 engine?

Do you have a case split apart in your life right now? (I am going to visit a pile of them on Saturday).

Is it possible that your photograph of the piston through the bore does not show the dump relief drilling into the camshaft gear well?
asiab3 wrote: I hereby wildly speculate (based on my two engines here) that the dual relief cases can use the tapered-grooved relief piston.
To what end? The groove, to be functional, has to have a channel for something to communicate with something else. I wouldn't doubt for a minute that someone convinced/sold someone else a grooved piston to stick in the relief bore, "hey, it says that it will help cool the oil!" and the bore has NO IDEA what do with it. It will serve as a normal relief piston without any ill effects.

My experiment with Chloe yesterday told me something valuable - the relief drainage can flow well enough to take away your oil pressure.
asiab3 wrote: I do not think the bevel/groove actually allows for more waste circulation in a dual relief T1 case. Could it be VW trying to get their money's worth out of the tooling for the grooved pistons?
I do not believe the grooved piston does anything for a dual relief case if the relief bore does not have the requisite passages to channel any oil flow when the relief valve is closed (warm) My AS-41 has a simple piston.

asiab3 wrote: This would mean the single relief case dump ports would have to be lined up with the spring extended (valve closed-ed)in a low-pressure scenario. Does anybody have one apart right now?
Robbie
You will see that the dump is always below the main gallery, but not beyond a groove's ability to let them communicate. They can only communicate with the piston pushed all the way down ( real cold) or all the way up, waste circulation On Those Engines So Designed. Did you see the AS-41 case half further up that had two holes in the camshaft gear well? One was the hydrostatic lock preventer vent hole, and the other sure did look like a dump . . .
Image

** It may be difficult to see the port in a shiny bore! **
Image
2019 Annual Funding Drive!
https://itinerant-air-cooled.com/viewto ... 41&t=13755

BobD - 1978 Bus . . . . . . . . . . .111,145 miles
Chloe - 1970 bus . . . . . . . . . . . 206,812 miles
Naranja - 1977 Westfalia . . . . . 94,655 miles
Pluck - 1973 Squareback . . . . . 55,510 miles
Alexus - 1991 Lexus LS400 . . . 76,560 miles

User avatar
asiab3
IAC Addict!
Location: San Diego, CA
Contact:
Status: Offline

Re: Lubrication Notes

Post by asiab3 » Tue Mar 03, 2015 10:23 pm

Amskeptic wrote:
asiab3 wrote: My old engine was a UO (1970 Type 3 1600) duel relief case and it had the grooved relief piston linked above. However, the groove in the piston does not line up with any dump ports in the old case, or the AS41 Brazilian case I'm building on currently. The groove lines up only with the main bearing gallery and plug, (picture below.)
Robbie, you "duel" to the death, a dual relief crankcase, however, is just a dual relief crankcase.
Is it possible that it was a retrofitted relief valve in your old 1600 engine?
Do you have a case split apart in your life right now? (I am going to visit a pile of them on Saturday).
Is it possible that your photograph of the piston through the bore does not show the dump relief drilling into the camshaft gear well?
Duly noting the dueling duals, I have two dual relief cases in front of me; one (shown below) has drilled gallery plugs. Both are identical in every way except head stud size.

Color key:
Blue = main bearing oil gallery
Purple = piston with spring fully extended (I drew these too high, apologies.)
Red = my only dump port
Green = piston with spring fully compressed

Image

Image
Amskeptic wrote:
asiab3 wrote: I hereby wildly speculate (based on my two engines here) that the dual relief cases can use the tapered-grooved relief piston.
To what end? The groove, to be functional, has to be a channel for something to communicate with something else. I wouldn't doubt for a minute that someone convinced/sold someone else a grooved piston to stick in the relief bore, "hey, it says that it will help cool the oil!" and the bore has NO IDEA what do with it. It will serve as a normal relief piston without any ill effects.
This is likely what happened. Looking around, that's all I see for sale. To me, in these cases, there is no way the grooved piston could function any way other than like a standard piston.
1969 bus, "Buddy."
113k miles with me.
292k miles on Earth.

kreemoweet
Getting Hooked!
Location: Seattle
Status: Offline

Re: Lubrication Notes

Post by kreemoweet » Wed Mar 04, 2015 1:16 am

So, it seems my conjecture than no Type I dual-relief case had a dump port still stands.

I found further evidence on my other conjecture, that the groove on relief pistons used by VW in single-relief cases do not communicate with
the dump port when in the fully closed position: the nice big engine cutaway drawing at the beginning of the Engine chapter of the '66-'69 Type I
Bentley manual. There, it is clearly seen that the dump port is well below the groove in the relief valve piston. It just wouldn't do, when your oil warning
light is flickering as you idle at a stoplight with a hot engine, to have that precious oil wasting away, unused, back to the sump!

Still wondering about those mysterious slopey-grooved relief pistons. Where do they come from? Are they even VW parts? Did VW even install them in any
air-cooled vehicles? Vanagons?

When VW brought forth those (straight-) grooved pistons into the world, they wrote: "As this measure [ed: replacing standard piston with grooved piston] reduces the
engine oil temperature over the entire range, the modified piston must not be installed in countries with temperate or cold climates." They soon reversed
themselves, and began using them in lieu of the solid pistons in all their single-relief cases, as near as I can tell. Do you suppose there was an accompanying change
at the same time in the piston spring specs, or elsewhere? I find no trace of it.
asiab3 wrote: ... nice big clear photos of engine cases with colored lines ...
I believe the purple areas are too low, as the top of the piston will be up against it's seat, which is about 7 mm above the main oil gallery. The green areas are WAY
too low, as the piston spring goes solid when the (grooved) piston top is only a few mm below the main oil gallery.

The use of grooved pistons (straight or sloped) in a dual-relief case with no dump port will not work the same as a solid piston, as the extra length of the grooved
pistons will significantly change (increase) the spring pressure under all operating conditions. Whether that be good, bad, or indifferent I can not say.

Jivermo
IAC Addict!
Status: Offline

Re: Lubrication Notes

Post by Jivermo » Wed Mar 04, 2015 5:04 am

After reading this discussion more than a couple of times, my head is spinning. More questions arise than answers.

Image

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: Lubrication Notes

Post by Amskeptic » Wed Mar 04, 2015 10:24 am

kreemoweet wrote:So, it seems my conjecture than no Type I dual-relief case had a dump port still stands.

I found further evidence on my other conjecture, that the groove on relief pistons used by VW in single-relief cases do not communicate with
the dump port when in the fully closed position: the nice big engine cutaway drawing at the beginning of the Engine chapter of the '66-'69 Type I Bentley manual. There, it is clearly seen that the dump port is well below the groove in the relief valve piston. It just wouldn't do, when your oil warning light is flickering as you idle at a stoplight with a hot engine, to have that precious oil wasting away, unused, back to the sump!
I will defer to your observation re the dual relief . . . until I get back from my inspection trip in Saturday.


Note that Type 4 dump looks totally different than above Type 1. It has two big channels below the main bearing gallery to help with cold oil removal.

Image

However, closed relief valve with the 1500 grooved plunger that causes " precious oil wasting away, unused (but cooled!-ed) , back to the sump!" is an established factory trick for reducing oil temperatures.
If you have a flickering light at (actually, above-ed ) hot idle, then you have either excessively opened up bearing clearances or a pump that has lost efficiency, or some other stupid problem that must not be "cured" by screwing around with this elegant simple system. This is the cold (haw haw) hard truth.

a) VW would not say, "well, just block off the dump port with a solid piston."
b) They would say, " check all bearing clearances and pump clearances,"
c) then WE would say, "no, that is too much work, I am putting in a bigger pump." The Road To Hell and all.
kreemoweet wrote: Still wondering about those mysterious slopey-grooved relief pistons. Where do they come from? Are they even VW parts? Did VW even install them in any
air-cooled vehicles? Vanagons?
I think those are a Gene Berg or other tweak.
kreemoweet wrote: When VW brought forth those (straight-) grooved pistons into the world, they wrote: "As this measure [ed: replacing standard piston with grooved piston] reduces the engine oil temperature over the entire range, the modified piston must not be installed in countries with temperate or cold climates." They soon reversed themselves, and began using them in lieu of the solid pistons in all their single-relief cases, as near as I can tell. Do you suppose there was an accompanying change at the same time in the piston spring specs, or elsewhere? I find no trace of it.
Do you have any sort of verification that the grooved plungers were de facto replacements?
Longer pistons may come from different sources. Porsche switched to longer pistons to reduce wear in the bore, others may have used the piston as a "spring strengthening measure". We had a spring length/strength change in 1970 with the introduction of the dual relief/10mm galleries.

Look at picture of the Type 1 case half. The lower hole is the "vent", the upper hole *should* be just under the main gallery. I will get back to youse all.
(03/04/15 : I got to see a single port relief dump port up close . . . slight dip offset from main gallery as drawn. It would flow with a grooved piston)

Image
kreemoweet wrote: The use of grooved pistons (straight or sloped) in a dual-relief case with no dump port will not work the same as a solid piston, as the extra length of the grooved pistons will significantly change (increase) the spring pressure under all operating conditions. Whether that be good, bad, or indifferent I can not say.
It will work precisely the same as a solid piston since all the work is at the top of the piston.

(extra credit essay)
So after reading my article, what would you observe with a longer piston? The actual physics.
Colin

(keep it up folks, I will modify the article after a thorough critique and go post links on the Shop Talk Forum and Samba engine forums. Then let the fun begin . . . )
2019 Annual Funding Drive!
https://itinerant-air-cooled.com/viewto ... 41&t=13755

BobD - 1978 Bus . . . . . . . . . . .111,145 miles
Chloe - 1970 bus . . . . . . . . . . . 206,812 miles
Naranja - 1977 Westfalia . . . . . 94,655 miles
Pluck - 1973 Squareback . . . . . 55,510 miles
Alexus - 1991 Lexus LS400 . . . 76,560 miles

User avatar
asiab3
IAC Addict!
Location: San Diego, CA
Contact:
Status: Offline

Re: Lubrication Notes

Post by asiab3 » Wed Mar 04, 2015 11:28 pm

Amskeptic wrote: (extra credit essay)
So after reading my article, what would you observe with a longer piston? The actual physics.
1. If two pistons with identical horizontal dimensions but different lengths were used in the same bore, the longer piston would wear less due to it's ability to stay straighter in the bore.

2. If the same spring was utilized for the two above pistons, the longer piston would require a higher dump hole.

3. Would the spring bottom out with a longer piston and need to be changed accordingly?

Robbie
1969 bus, "Buddy."
113k miles with me.
292k miles on Earth.

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: Lubrication Notes

Post by Amskeptic » Thu Mar 05, 2015 11:25 am

asiab3 wrote:
Amskeptic wrote: (extra credit essay)
So after reading my article, what would you observe with a longer piston? The actual physics.
1. If two pistons with identical horizontal dimensions but different lengths were used in the same bore, the longer piston would wear less due to it's ability to stay straighter in the bore.

2. If the same spring was utilized for the two above pistons, the longer piston would require a higher dump hole.

3. Would the spring bottom out with a longer piston and need to be changed accordingly?

Robbie
So, out in the real world, if someone were persuaded to stick in a bigger piston, they could:
a) upon cold start, blow up of their cooler and or gallery plugs because the piston could not clear the dump hole with spring bind.
b) resultant heavier spring pressure would also cause early cut-in of the oil cooler.

I'll get back to you on dual relief dump relief.
Colin
2019 Annual Funding Drive!
https://itinerant-air-cooled.com/viewto ... 41&t=13755

BobD - 1978 Bus . . . . . . . . . . .111,145 miles
Chloe - 1970 bus . . . . . . . . . . . 206,812 miles
Naranja - 1977 Westfalia . . . . . 94,655 miles
Pluck - 1973 Squareback . . . . . 55,510 miles
Alexus - 1991 Lexus LS400 . . . 76,560 miles

Post Reply