Doin' good so far Hap!

"I may consider idiot warning lights now, as we get older, we need all the aids we can get"
This is not, as some would think, because our eyes are going, or hearing, or attention diodes. Rather, it is because your brain is finally starting to run the show instead of your balls.

I've been preaching the evils of excessive oil pressure for years, but it was casual "might gain a little", especially long term durability, not "avoid catastrophic failure." New ball game in this story!

Skew gears suck! Very poor design to carry heavy constant loads. (some Mercedes have a cute little single row chain driving the oil pump) Requires exquisite balancing of materials and lubrication, which should be hypoid EP oil but ain't. Matching materials in such gears is black art, though I'm sure computers help nowaday.
Bronze gears are good for the lube requirements of these things, but they suck for load carrying. The drive was designed for steel gears to some spec, and bronze ain't gonna do it at heavy load. Especially if you do not know what of 10,000 bronze alloys are being used.
So, a factor is gear material and heat treat and finishing which could be very different with current parts than earlier ones. People nitriding or titanium ion coating cams, maybe gears (don't know?) - the old ones did not have that and hypoid/skew/worm gears require matching material pairs. Attention to these plus possible discussion on modern antifriction coatings seems in order.
I do think that this is probably the source of the current problems.

Bad repro gaskets are inexcusable, Correct machining and the tiniest bit of LocTite PlastiGasket or whatever they've changed the name to solves. There is, after all, no gasket at the cover/body interface so lap/scrape the parts together, or just make a correct gasket - Ink blotter paper ain't. How about .002 soft stainless tool wrap, or copper foil, widely available and could be water jet cut for a hell of a lot less than blown motors.

Now that you have a pump in a definite location, and after you measure all the parts outside, assemble as Dave said and measure ALL the assembled clearances, including the drive spindle.
While doing that the machining or casting errors will show up. Then measure drive torque.

Now you have eliminated all spurious loads, what are the required ones?
MGB was designed for 70psi relief pressure, but that is not to say it needs to or should be at 70 once hot and running. Comp tune book says "between 70 and 80 and up to 100 is a good pressure" which I think is old school claptrap; and then goes on "dropping to 30-40 is satisfactory" No rpm numbers.
"50-80psi = normal running pressure" Clearly the 80 is cold with the relief spec'ed at 70, and the 50 must be hot driving, say 3000.
There are arguments that max bearing loads could be either at low speed or high, but pump pressure is most important at low speed - it will automatically increase with engine speed. At high speed hydrodynamic factors predominate so long as oil is actually supplied, and the bearing in effect makes its own pressure, increasing load capacity as speed increases.

Here's the real kicker - you are not measuring pump pressure, rather delivered and maintained pressure after the pump/relief/main delivery passage.
If you guys have, as I think is the case, optimized passages from the pressure measuring point to the bearings, then you have way more pressure than you need even there. 30 psi would likely do, but I don't want to cut it too close.

Edit/Add on final reread: I think it was said the the convoluted #4 main passage causes trouble? This could be a cavitation result from TOO MUCH flow at high speed.

The actual pump pressure and flow is determined by pump size and speed, passage size, relief size and setting (two different things). Positive displacement pumps deliver in a linear fashion by speed, so whatever happens at 2500 doubles at 5000 and triples at 7500. The power requirement will increase by that, plus a factor for a resistance/loss coefficient that probably is logarithmic increasing. Were a smart feller to measure actual pump output pressure (I think there are ports already in the block!), and calculate flow, he could get a real "power eaten" figure, and I guess you might know why them leetle gears is cryin' and dyin'. Good news is you could then rectify this mess and feed the horsepoppers to the rearmost tires!
Another point on the gear failure is that if the system is running against relief pressure all the time with unknown loads, the poor gears never get to rest and cool off. Very different to an engine that runs below relief except for rare excursions to rod heaven. So in addition to total load over time, there is a time dependent temperature factor.

So, I see mo point in oil pressures above 60psi, and that should generally be below the set relief pressure to ensure that large, hidden, useless to detrimental flows/pressures are not happening. 45-50 hot at 3000 should be adequate if you are not Aunt Estelle going to bingo in 4th gear at 25mph. And no surprise if you could shorten the pump gears to reduce parasitic loss still more. Might make gear pumps with more teeth as well, rather than the rotors -
I haven't thought about pulsations in rotor pumps, but I'm sure the data is in hydraulics texts.
,
Having wrote all that and shot hell out of another night, I thought to look up some "experts"; testimony follows:

random info:
Jenkins said the SBC gained 1% power when oil temps were held at 220 vs 180.That would be 4-8 hp, probably near 1hp/1000rpm if you figure the high hp engines were turning 8000.. If you figure that some major part of that was used pumping oil, that's a lot of gear load. And, that is only the excess from heavier oil, not nearly the whole pump load.
And "if we race an engine, the oil temp IS AT 220, not 200 or 225"
He also set hot oil pressure at 55-60, which might give cold start pressures around 80+.
Same pressures on BBC - On SAE 20 or 30 oil
And on BBC he ran standard 1.14" gears, not HP 1.3 gears pumps."the bigger the pump, the more hp it takes to drive it, and the more oil it takes to feed it, and the more you have to get rid of when it blows the bypass open - and the better the chance of cavitation."
"They purposely built the BBC pump with more teeth on the gears to reduce pulsations transmitted to the distributor drive gears. Previous (SBC) pumps(with fewer teeth) caused spark scatter and gear breakage at high rpm." (same gear drives pump & dist)

Vizard A series: "...release valve set to give 80, 90, even 100psi. Oil pressure this high does not protect the bearings any better than does oil pressure around 60-65psi. What it can do though is cause more hp to be absorbed in driving the oil pump in the first instance, and secondly it can unnecessarily raise the oil temperature."


FRM

Flecther sounds like a buisness opportunity for you. Sounds like you got it all fiquired out, now all for you to do is make and sell to to likes of us

---

Hap Waldrop
MGexperience Supporting Vendor
Make Sure to check our promotions/product offerings in the Vendor Deals Forum here.

Fletcher, thanx for posting your thoughts on MGB oiling and I can, in principle, agree with most of it.

A little history is in order. Back in the '60s when these cars were first being raced here, max engine speeds were probably not over 6500 and 3-main engines will do fine at a lower oil pressure than perhaps a 5-main. Before "Cal-Club" became part of SCCA, they allowed prod car racers to "prodify" their engines far beyond what SCCA allowed and beyond, I suspect, what the factory race teams were doing as well. This of course started pushing redlines higher etc. And of course traditional thinking wanted higher oil pressures with higher viscosity oils.

Followed closely by failure of the pump drive spindle gear. Higher RPM and higher pump load being the cause. That's when Huffaker had bronze gears made. The benefit was, when you wore the spindle gear it didn't destroy the cam. BTW, It's common to use a bronze gear on SBC(Chevy) pump and dizzy drives, I believe for the same reason. Monkey see monkey do! Of course Joe wanted to periodically sell you a new cam for more power, but not because the gear was shot.

I use the bronze gears today when I'm building an engine with a OE type oil pump for the same reason. By observation, while not as strong as steel, the bronze creates an initial wear surface without galling and greatly reduces unit tooth to tooth contact stress. These generally give very good life although it's a good idea to relace the gear when doing a refresh of the engine. It's my impresion that the bronze alloy used today is the same as that specified by Huffaker.

Of course if something cause excessive drag on the pump drive as I've alluded to earlier in this thread, Then you'll be SOL with OE steel or a bronze gear.

Regarding convoluted oil flow to #4 main, this bearing does not feed any oil to the rods so I think it gets as much oil as it needs to feed the crank through the 2 bearing oil holes.

The oil pump gasket spaces the pump and gears to provide needed end float and of course prevents sucking air with loss of prime. I have not compared compressed thickness of the Payen gasket to other gaskets although I think uncompressed it is thicker to start with.

Fletcher, it's not like I have not been thinking about this and how to not repeat the issue again.

The cam that was being used was nitrited cam, so that made the gear on the cam harder than it needed to be when compared to the gear on the pump, drive, but the distributor drive is fine, even with the cam gear being chewed up, so ponder that one.

These are the options I'm considering.

A. Replace the cam with a cam that has just hardened lobes, but not hardened gears, and use a another steel gear, least likely option.

B. Replace cam, use the bronze oil pump gear. I already seen one of these gears fail in short order, but others have used them for much longer periods, but this is really a bandaid on shotgun wound approach, if the bronze gear fails it will saves the cam, but if you don't see the low oil pressure quick enough, you could destroy the engine, and the gears, which are not cheap, needs to replaced with every engine refresh.

C. Use a different type of cam, with larger gear count and drive that matches it, the cams are rare to come by, but the drives are available. For the most part this has been a reliable choice for those who have used it.

D. Do what Dave did with his vintage car, use a single stage external oil pump, this would take some fabrication, but nothing beyond what I can do, it's alittle more expensive than the other choices but should pay for itself in the long run, and the gear and it's wear issues are taken completely out of play. This would also make for a more efficent windage tray in the oil pan, getting rid of the oil pump, drive and such,and I can put the pick tube in the oil pan wherever I want to.

---

Hap Waldrop
MGexperience Supporting Vendor
Make Sure to check our promotions/product offerings in the Vendor Deals Forum here.

Hap, one consideration on using the external wet sump pump. The one that I have used has a sensitivity to sticking the regulator with fine debris. This due to regulating unfiltered oil. The next one of these I do, I plan to modify the block so I can use the OE regulator in the block and lock down the pump regulator so that it only opens at a very high(100psi+) pressure. The OE regulator in the block will then be seeing filtered oil.

I don't think the bronze gear has a wear issue if the pump is free to turn when installed. I just freshened a race engine with 3 full seasons on it and the bronze gear was serviceable although worn enough that I replaced it.

BTW, the dizzy drive is so lightly loaded that I'm not surprised that it survived.

Hap... the cams for the Datsun (1600?) have a higher gear count, but it might be easier to find Atlantis than to get ahold of the cam and drive gears for the distributor and oil pump!

Dave-
Pretty much as I thought, and confirms my idea that this is an issue of design pushed past assumed limits. That Huffaker spec'd the bronze is a very good thing, so long as that is still what you get when you buy one. And I think the Chev ones were a factory item, not sure. Lots of engineering and the resources to find & use the right stuff.

So I was thinking in terms of dumping old school assumptions and trying to reduce the loads, as a way to avoid total redesign of the pump(s) and drive.

Here's a thought: Do you suppose the fully loaded gear has enough end thrust to add a significant drag at the thrust faces? The oil feed to the top end of the spindle is about 6 miles downstream from the pump, out of the centre cam bearing/low pressure gallery. The feed to the lower thrust face seems to be only that which leaks up the driveshaft from the pump.

Your "replace on rebuild" is good practice, if you actually catch the gear before failure, but it still means the gear is marginal. Hap's saving most of his engine is a miracle, and we do not want to use up the miracle allotment. My impression from your post and Hap's comments is that this formerly "solved" (by Huffaker) problem has popped up again, so what has changed?

Hap -
I have no doubt you are wracking your noodle trying to solve this. A few thoughts:
These gears require matched metal pairs plus correct lubrication, and hardness as such is not necessarily a big factor.I have the original papers from the early 50's re the puzzle of cam/tappet failure, and some on hypoid gears, and it ain't obvious!

Hardened gears is not the issue when you are actually coating them as well. A lot of these vapour phase or liquid cooking coatings are fundamentally crystalline and abrasive, despite the fact that they may give excellent EP characteristics. In some applications, the same coatings are used to increase friction, and if you want to reduce it, you have to polish or lap the coating. How the two materials interact with each other and the lube, and at what loads, is quirky.

Plan B with untreated cam gear and bronze pump gear <should> get you back to Huffaker. Do we really know what pressures and speeds they were using?

And the "wake up dummy" flashing lights and klaxon. My big truck did all that to tell you that it would shut the engine down in 15 (settable) seconds = Hawk warning system I think.

Dizzy drive is not likely to fail until the cam gear is so bad it grabs the dizzy one, as Dave suggests. But, the dizzy is still subject to cam and pump pulsations, as Jenkins found. At high rpm, they got mysterious problems, which turned out to be advance weights vibrating/bouncing against the solid stops - so they used very stiff springs at full advance and rubber stops as a final advance limit = cure.

A thought is to make a high pressure oil jet aimed at the gear interface upstream side.
Could be the same system feeding the piston squirters... Mazda has these set to operate only when system pressure is above 35psi.

And as above, I would be measuring pump pressure as close to the pump as possible. Has anybody ever enlarged the pump>>relief and relief>>drain passages to reduce the inevitable excess pump pressure?

And Hap-
If you refer to that gasket, I personally would just do away with it, and I have a time or two, because it was 3am and I had none. But if I spent the money to make 500 of them, I'd sell six, and you lot would say "I can get that made for $.42, so why should I pay $4.00? Besides that, I'd probably give them away because I had that pile on the shelf, and no chance of ever selling them. I am notoriously deficient in money producing ability, and greatly in awe of anybody who can make 3 bucks honestly!

FRM

Summat disappoeared someplace I think.
Some Mercedes have a cute little roller chain driving the pump off the crank nose. In idle moments, I have thought of building a whole chain drive dry sump system inside the pan, with only pipes coming out.
Or drive the pump off the front of the cam, like an old Hillborn FI system, or a flathead Ford dizzy.

FRM

Fletcher, if you want to see a nicely done single row oil pump chain drive, check out the 1996 and later GM 2.4L twin cam engine. BTW the chain also drives the counter rotating balance shafts and the pump drive loading dampens the gear rattle.

I'm pretty confident that the amount of bind in an MGB oil pump when tightened like I described earlier is the most likely cause of recent issues. Not much has changed in the area of the system since the late '60s except the afore mentioned gasket.

I'm sure when Hap puts things back together he'll repeat the experiment. Maybe Dick Moritz as well.

Dave-
Thanks for turning me on to that. I haven't kept up with modern developments, so I was not aware of these. Will need to try to find pitchers! I had thought that the whole Quad 4 family and children had been killed off - never came to me that they would bother to put those stupid covers on. Only one I ever saw was in a pile in a junkyard, head facing me. Talk about brake checks, thought I'd found an Offy. Sad part was it had about 1 sq ft airflow area in the side of the block!

FRM

Fletcher, the GM Quad4/Twin Cam engine family was used through 2000-01 IIRC. Many of the Quad4 engineers(myself included) were involved with the design of the Ecotech which replaced it.

Hi from Australia. Drive gear failures known here also but are 100% overcome by 1. using cam blanks that are centre drilled and drilled thru base of drive gear to feed oil pressure from rear cam bearing to the oil pump drive gear interface & 2. tuftriding the cam to harden the gear surface. Oil pump gears are used standard and unmodified. Hope this helps. John Stevenson.

Brilliant idea John, the gear is only getting its lubrication from the splash and at high RPM’s what little oil it gets is slung off causing premature wear. If the gears were submerge like a differential, their wouldn’t be an issue, force feeding it is one I can except. That’s why some of the guys were using bronze, because it is self lubricating by nature, why not keep the steel and feed it with oil under pressure, and then you have both worlds.



Edited 1 time(s). Last edit at 07/16/2012 07:34AM by BlueMax1.

Yep, that would help the sitiuaction. Ironicly APT gets A series cam blanks that are gun drilled the entire lenght of the cam, and has oiling holes at the cam lobes, and plugs on each end, sorta like the rocker arm shaft. If oiling is the issue than this would fix it. now I could source a steel MG B series cam blank from David Anton, do this modifcation to the cam, you wouldn't really even have to reharden the steel billet, it already the correct hardness, the way I see it, is you would gun drill from the back side of the cam, until you got to where the gear was located, then drill a oil hole at the gear, and feed hole from the cam journal, then send off the cam to to the grinder that has my master lobes for my race cam. Only one catch, David is out of the country until fall, and I really would not want to do this mod to a stock iron cam.


At the end of the day, the external oil pump, while a expensive option sounds like even a more interesting option because it completely take this out of play.

---

Hap Waldrop
MGexperience Supporting Vendor
Make Sure to check our promotions/product offerings in the Vendor Deals Forum here.

Quoting my own post #27:

"A thought is to make a high pressure oil jet aimed at the gear interface upstream side.
Could be the same system feeding the piston squirters... Mazda has these set to operate only when system pressure is above 35psi."

Access/Space in there is a little tight, but a few holes drilled and a bit of tubing could bleed oil off the pump delivery. Possibly straight out of the pump, no block mods. True it is not filtered, but that seems a small issue vs, shrapnel from failed gears.

FRM

2X Fletcher, NASCAR boys do that stuff all the time. Only if you don’t have any issues with interference with crank or rods the oil galley is right their next to it. That would work perfect, bathing the gear with its on constant supply of oil. You wouldn’t have to worry of oil starvation to the gear and you wouldn’t have to center bore the camshaft. It’s not as clean looking, but very efficient!

Also their needs to be enough meat in the block at the galley aera to thread into, if to thin it may interfere by choking the galley down enabling oil to be supplyed through the engine adequately.



Edited 1 time(s). Last edit at 07/16/2012 02:59PM by BlueMax1.

My cam grinder here gets his MGB blanks from USA already configured with this oil drilling.It is common practice here. He is insistant that tuftriding is also important. John.

Alan, and Fletcher, I got a very intresting email today from a fella, that I'm sure read this stuff we write, but not sure he ever posted here, and he races a SCCA MGB, he shared alot of info with me, about how to increase oil flow to the gear, to be honest with you, on Monday with all the emails and phone calls I get over the weekend, I read it, but not absorbed it all just yet, I very much appreciate him taking the time to write that, and send it to me, he has some good idea,of how to get more oil to the gear, and has done this on his race car with sucess. So I've got take some time and read it all again, and study it all, and plan on trying his suggestions, and will glad to share them with the group,and give credit, where credit is due, as I know more about what I'm talking about.


John, I get what you are saying in whole, and I think that is good idea as well. Greenville, SC is avery resourceful town, beleive it, or not, I have place here in town that ION nitrites, and they don't mind samll batch jobs like me, so that would be very do-able.

---

Hap Waldrop
MGexperience Supporting Vendor
Make Sure to check our promotions/product offerings in the Vendor Deals Forum here.

http://www.mgcars.org.uk/peterburgess/cams.html

Interesting info on Piper cams and oiling of skew gear.

Well how about that. The link gives you a photo of what I was describing. It works!!! John

Also Google Tuftride Heat treatment and find article from lotus Seven Club on Surface hardening of Steel. It comes up first. For the relatively small cost ( about $100 here) well worth considering. Dont use bronze gears!! John Stevenson



Edited 1 time(s). Last edit at 07/16/2012 11:55PM by jdstevo1.