Oil pump gear material 1

[yoshimitsuspeed]

I am talking to a company in China that makes oil pumps for the motors I specialize in. I am still feeling out the idea of using their products and got a handful to inspect and test so we will see how that goes.
One of the main things that interested me would be getting them made with high strength metal gears. The stock gears are sintered metal and are known to fail on high RPM builds. There is enough middle ground between what the stock pump can handle and the need or budget to go full dry sump that there is a market for high strength gears.
Here is an example of some that I currently sell.

http://www.todaracingusa.com/catalog/pc/viewPrd.asp?idproduct=655&idcategory=90

I really thought these were made of chromoly but I might have been mistaken as I can't find anything about that now so perhaps it's another alloy.
When I thought they were made of chromoly I asked the company about having them make chromoly gears and they said it would wear the aluminum housing much faster. This seemed odd to me but it did get me wondering about what the best material would be and if there would be any other concerns like that. I don't really see how something like chromoly would cause it to wear faster but this level of metal engineering is well beyond me.
Does anyone have any thoughts on what desireable alloys might be and if there would be any concerns of using these other alloys?

If anyone is curious about the failures or has any input on that I would be very interested to hear that as well.
At extended high RPMs they have been known to break. I don't know if this is due to force, wear, or harmonics or possibly some combination.
As the gears get older they definitely get some pitting in them and in the high RPM motors this pitting is more severe. Some of this may be due to the sintered metal not being held together very well but I suspect that it at least partially has to do with cavitation though I have heard that the high strength aftermarket gears don't really get this or at least not as significantly. Unfortunately I haven't gotten to inspect used ones yet. At some point I would like to do a little research to test pressure at the inlet of the pump as well as play with some porting to see if it's possible to reduce pressure drop on the inlet side. If there is significant pressure drop this would definitely contribute to more cavitation.

I have also wondered about the interface between the gear and the crank. This is not a terribly tight fit and I have wondered if it could actually clatter a little bit. I have wondered if something like a very thin delrin or similar sleeve could prevent this from happening but then I have wondered if it's better that the gear has a little room to float and if attaching it to the crank more rigidly could cause problems.

 

[tbuelna]

The gerotor oil pressure pump elements you linked are made using sintered metal for production applications because it provides the desired balance of cost, durability and performance. Making the same gerotor oil pump elements from wrought CrMo alloy steel would provide better reliability and fatigue life, but would not likely affect the "wear rate" of the associated aluminum housing surfaces.

 

[yoshimitsuspeed]

Thanks tbuelna that's pretty much exactly what I was thinking.
Definitely interested in anyone else's input on this. Especially if anyone had any thoughts on specific alloy, temper or hardness. I am thinking hardened would be good but too hard might lead to it being too brittle. Or would that be of minimal concern in this application?

 

[MikeHalloran]

WRT rattling on the crank, a light coat of Loctite Pipe Sealant with Teflon would provide a customized sorta-sleeve that's also removable.



Mike Halloran
Pembroke Pines, FL, USA

 

[yoshimitsuspeed]

Mike I don't remember for sure but that pipe sealant fully cures correct? So you are suggesting something that would lightly bond the oil pump to the crank with something that had some dampening characteristic? It's an interesting idea but again since I don't know for sure the cause of the stress it's hard to know if that would make it better or worse tying the two together. Since there is a little slop it might also be possible to bond it slightly off center.

 

[gruntguru]

One downside of "hard" gears is tolerance of debris. Softer gears will allow a small amount of "imbedding" of foreighn objects. Hard gears may jam and fail the drive system.

je suis charlie

 

[tbuelna]

Looking at the picture of the failed pump gear in the OP it has fractures propagating from the tooth root fillets. When you consider that this gerotor pump is basically an external gear driving an internal gear, this type of failure would usually indicate excessive flank loads on the external gear teeth. The two most likely causes of excessive tooth flank loads on the drive gear would be high oil discharge pressure or seizing of the internal (driven) pump gear.

High oil pump discharge pressures could be caused by running at high rpm with very cold oil, or having a pressure relief valve that is not functioning properly. Regardless, this is a condition that would be easy to detect.

Instead, what I think is more likely the cause of your problem is binding/seizing of the gerotor gears within the pump housing. The external (driver) gerotor gear is cantilever mounted on the crank snout, with both a cam drive pulley and an accessory drive pulley mounted to the crank snout forward of the pump. The stock gerotor gears are designed with a close tolerance running fit inside the pump housing to minimize leakage and maximize efficiency. The increased drive pulley loads from operating at higher power/rpm may be causing enough bending/displacement at the crank snout to create interference between the gerotor gears and the pump housing. If this is indeed the cause of your problem, it could be addressed by increasing the running clearance at the side faces of both gears, and at the internal gear OD, just enough to prevent binding. This will reduce your pump efficiency slightly, but it shouldn't be a concern with a high-performance application.

 

[yoshimitsuspeed]

On these the internal gear is driven by the crank and the external gear is driven by the internal gear.
The newer designs use the round teeth instead of the square which are supposed to be better but the square was just the first pic I found of a failed one.
Sometimes the inner gear is broken, sometimes both but I have never seen an outer gear broken with a healthy inner gear so I would suspect that generally it's the inner gear failing and sometimes that destroys the outer gear as it comes apart.
I believe this is a Viper pump.

Another Dodge

Festiva I believe

It seems to me a very likely possibility is that the force of the crank trying to spin the gears at high RPM may just pry the gear apart but I have wondered if other things could contribute to that like harmonics or cavitation creating a hammering effect of the crank on the oil pump gear. There are definitely signs of some pretty serious forces on some of these flats on the gear that the crank presses against.

Whatever the case these upgraded gears definitely seem to resolve the issue.

I don't know how the OEM material compares in strength to anything else but I'm assuming most any steel would be better. I figure 4140 or similar should be pretty good but I have been looking at 4150 for it's high wear resistance. Still wondering if it would be better to go high hardness or a little softer. gruntguru has a fair point about imbedability but gears like this will generally be run on more performance motors with more frequent oil changes and more monitoring and inspections. But who knows if harder would be of much benefit. Afterall a steel gear will outlast a number of aluminum housings unless something goes through it.
The 4150 on Mcmaster come in B90 Rockwell but can be hardened to C63. Maybe somewhere in the middle? Like C30-C40?

 

[MikeHalloran]

The witness marks in that last picture suggest a disadvantageous pressure angle in the drive details.

Regardless, I'm thinking it sort of smells like the discharge pressure is rising too high.
Are you shimming the relief valve to increase oil pressure, or running super tight clearances on the engine bearings, or doing anything else that would increase the load on the oil pump drive beyond what a totally stock engine would see?



Mike Halloran
Pembroke Pines, FL, USA

 

[yoshimitsuspeed]

Yeah I just measured a crank and the flats that drive the pump are about .5mm smaller than the flats on the pump. This does allow it to twist and really focus pressure on the corner instead of pressing flat against the whole surface. It will also allow it to hammer back and forth especially if compounded with any torsional harmonics.
I do think that as far as this issue is concerned tightening up that gap would help a lot. Either machining the oil pump flats narrower to make a tighter fit on the crank or machining the pump a little bigger to allow for something like a Delrin sleeve with a tight fit that would even pressure and absorb some harmonic.
My concern with either and I suspect why it's so loose is to account for any concentricity issues between the pump and the crank. If i were to eliminate that freedom of movement and there were any variance in concentricity this could apply new undesired forces to components. If that could be shown not to be an issue I do like the idea of having something like a delrin sleeve.


Now I will say that I haven't experienced this firsthand so I haven't gotten to study it first hand. I also don't think many others have spent a lot of time trying to figure out why or how. Someone initially came up with these stronger gears and if there was any concern people would run them and they stopped having issues. So it may be possible to make the stock gears hold up better but why risk it? At the same time if there are things contributing to the failure that could be addressed in the stronger gear that would be great.
Personally I am not a fan of raising the oil pressure but others do so I am sure at least some of these had shimmed or stiffer bypass springs. Even if they didn't there would be some increase in pressure as the spring opens more and resists more as the volume going through the bypass increases. Even if the pressure stayed the same the force applied to the gears would still increase with RPM due to friction and flow limits of the pump. If I had to guess I would bet it's a combination of these things. The play in the crank to gear interface allowing a sharper point of contact, possibly harmonic hammering and increased force on the gears pushing outward. Loose play, chattering, or jamming between the two gears could be a contributing factor as well. I assume this might become somewhat likely if the pump started wearing toward the outside of spec. I would assume that with the pump we would want to go tighter tolerances to perform better at higher RPM and flow but this is just a hunch as well.
If I did end up getting these stronger gears made my thought at the moment would be to make them at the very tight end of service manual spec between the housing and the gears and between the gears themselves and maybe half the gap between the crank and the gear. Leave a little freedom of movement but much less.
Another thought I had was possibly getting them made on the very tight side of spec and then using diamond compound in the pump and spinning it to lap it lightly. I am wondering if that would be worth the extra step to make sure everything is mated up all nice and perfect.

 

[MikeHalloran]

Now you've got me wondering if people are running looser than stock crank bearings, forcing the pump to take the load from a main bearing, in which case reducing clearances would be a bad idea.



Mike Halloran
Pembroke Pines, FL, USA

 

[gruntguru]

If the crank connects to the gear with multiple flats with the pressure angle shown, the gear will either:

- be forced to be concentric with the crank (all flats driving) or
- the contact line will move (as the radius changes) and will transfer from one flat to the next (single flat driving in the region of minimum main bearing oil film thickness)

If clearance needs to be increased, it would probably make most sense to reduce the OD of the outer gear. This is a leakage path of course but a fairly long one so oil pressure loss may be acceptable. I assume we are only talking a couple of thousandths.

je suis charlie

 

[yoshimitsuspeed]

Many of these higher reving motors would be running bearings on the loose side but we would be talking about a couple hundredths of a mm.
I would really hope that the pump would be within .1mm of concentric with the crank. Hopefully a lot less than that. So that plus potential crank flex, it seems like .15mm-.2mm should be more than enough gap between the crank and the pump gear.

Or use a thicker sleeve of something a little more forgiving that could absorb a little more movement. It seems like this could possibly be better for the crank and the pump but maybe it could cause issues of it's own.

 

[tbuelna]

The Toyota 4AG crank has two diametrically opposed flats (item C in the image below) that the pump gear fits onto.

The newer pump gears have bore profiles with mating flats that have been relieved in the center section so that they only contact at the outer edges. This results in a more consistent and predictable contact condition with the crank flats. If there is even a tiny amount of clearance between the mating surfaces with this type of coupling, it will tend to edge load on the two opposing corners that contact in the driven direction.

 

[yoshimitsuspeed]

Yeah you can definitely see deformation on both the crank and the crank gear at the contact point where it flattens out until there is a couple mm of contact area and deformed metal. This is looking at a crank that came out of a normal car, not a race motor. This deformation increases the contact patch but would also put more distance between the two components. If there is constant pressure this increased distance may not be an issue but if there is any harmonic hammering then more room could allow for greater impact force.

Does anyone have an opinion on using an insert? Option 1 might be to make the oil pump drive surfaces bigger to allow a sleeve insert to go between the crank and the gear.
Another cheaper easier and more conservative option might be to have an insert with a little give. Maybe Delrin, Teflon or high durometer urethane that just filled the space between the drive flats. Either a smaller one like the first pic that just went between the drive flats or a larger one that filled all voids. The crank has a second set of flats 90 deg off.
The first design would be real easy to make and if it was a tight fit between the crank and gear it seems like it would help spread out the drive forces a little and provide a little damping.

 

[yoshimitsuspeed]

I guess I can only upload one file per post. Here is the bigger one.
This would spread those drive forces over a lot more area but would also cost a lot more to make. The thin areas may also be too thin to make practically or work properly. If those were removed the second set of flats on the sides probably wouldn't do much.
If I was getting my own gears made I could get the inside profile made differently to work better with a design more like this.

I am thinking since no one has really had issue with the upgraded steel design it might be a decent first step to get these made with a little tighter interface between the crank and the gear and then possibly offer that small insert as an option. I am having a hard time believing something like that would actually cause a problem or make a problem worse but it's hard to know for sure knowing what I know and with something like oil pumps it would really suck to find out the hard way that a new design had an unexpected flaw.

 

[MikeHalloran]

Urethane won't survive normal engine oil temperatures, so that's a nonstarter.

Delrin has been used for timing chain sprockets, but there was a pretty good market for ferrous replacements, too, and given that the stresses on the drive flats are high enough to deform steel, I don't think Delrin is a good choice for that, either.

I'd be inclined to make replacement gears with the same interrupted flat that appeared on the more recent stock gears, and not tighten up the clearances, for pretty much the same reasons the factory did it.



Mike Halloran
Pembroke Pines, FL, USA

 

[yoshimitsuspeed]

If I used an insert like the ones above the hope would be that it would help dissipate some if there was any harmonic hammering on the gear while spreading out the forces a little bit. The hope would be for the crank and gear to make metal on metal contact before permanently deforming the insert so it continued to help a little while the primary load would still be carried by the metal on metal contact.
I would assume that the wear on the crank and the oil pump is due to the focused pressure of the sharp crank edge on the gear flat, possibly combined with any rubbing or harmonic impact. In the idea of using an insert that went inbetween the crank and the drive flats would be that it would create a tighter fit with much more surface area. So while the sharp crank edge and virtually no surface area pressing on one spot may be enough to deform steel the hope would be that a 6mm x 14mm of something like delrin would hold up much better because of the added surface area and slight harmonic absorption.

I have some theories as to why the factory might leave such loose tolerances between the crank and the gear but if you have any theories or actually know why they would then I wold love to hear them.