Oil issues: Dry Sump, Accusump for road racing?

[biknman]

I see the Daily kit out there anybody consider the Roger Clark Motorsports RCM kit if you catch it when the dollar is strong against the pound it can be a sub $3k kit. I’ve been on some UK Suby Racing forums both for TT Hillimb and Rally and those running the kits have had great success. Ref:

http://www.rallispec.com/eng_opu_rcm2668.html

03D27744-8CB5-4C7E-A07B-CB8D38794AA3.png

http://rogerclarkmotorsport.co.uk/up...5c011d7df1.pdf

21682407-47AA-41D9-A5E6-BA47D52359DC.png

[Scargo]

I've seen that many times and I priced it once and though it should have been cheaper in the US relative to RCM's price, but it wasn't at all. I don't think they will play that game.
I don't see anything magical about their pan. Water pump is standard Subaru part I've been using.
I DO like the undermount design. I have to ask, "where is the pressure regulator"? It is not mentioned. The oil pump cover plate is no big deal. A stock pump can be modified.
I know you are a very capable person. Why can't other pumps be made to sit next to an ARE pan?
I don't have a machine shop anymore and am not good at CAD. Could hire it done. Do have TIG. Think I could piece together something as good as anyone's, but I still like Dailey's system...

[biknman]

Yeah diminishing returns considered seems like a pretty well tested and thought out kit. I like the under mount for low GC and the reduction of hoses the RCM ZEN and Daily kits support. And it goes along with my thoery of K.I.S.S. And keeping it simplistic.

BTW they also sale the oil cover f/uw pump oil feed or filter feed: http://rogerclarkmotorsport.co.uk/st...ver-cover-feed. Along with some other bits and prices for doing ones own dry sump config.

[biknman]

FYI here some links from users of the RCM kits and others for reference I’ve been following for quit some time, seems like the folks across the pond have a lot more history with dry sump scoobys for Motorsports to draw from:

https://www.scoobynet.com/general-te...longevity.html

https://www.scoobynet.com/members-ga...it-fitted.html

http://bbs.22b.com/forums/showthread...y-Sump-systems

[Sgt.Gator]

The RCM is an excellent system. I see it runs off the timing belt too, so it should fit ok, other than a custom exhaust header, which you have to include in the overall package price.

[Sgt.Gator]

I have to ask, "where is the pressure regulator"? It is not mentioned.

It's in the PDF: Preload screw on the side of the pump.

[Zach34]

I've seen that many times and I priced it once and though it should have been cheaper in the US relative to RCM's price, but it wasn't at all. I don't think they will play that game.
I don't see anything magical about their pan. Water pump is standard Subaru part I've been using.
I DO like the undermount design. I have to ask, "where is the pressure regulator"? It is not mentioned. The oil pump cover plate is no big deal. A stock pump can be modified.
I know you are a very capable person. Why can't other pumps be made to sit next to an ARE pan?
I don't have a machine shop anymore and am not good at CAD. Could hire it done. Do have TIG. Think I could piece together something as good as anyone's, but I still like Dailey's system...

I've been dabbling in CNC with a 3-axis mill. I've become relatively comfortable with CAD/CAM. I'm confident I can machine things like that RCM front cover and oil pan. I intend to do my own DS system. The Cosworth, Dailey, RCM, and ARE systems all have their plusses and minuses. I think I can combine them. My problem is mostly time. I hope to start work on the major parts in January. The only thing I can't make are the pumps, so I will have to adapt someone else's bare pumps to my design.

[Sgt.Gator]

Zach if you make a Cosworth/Aviaid style pan I have a suggestion or two... I worked on a design but scrapped the project because I don't have a CNC machine and the cost to produce outsourced vs the size of the market just wasn't economically feasible. But here's a couple of thoughts for you:

The Cosworth pan uses an O-Ring instead of Three Bond or Permatex to seal it to the underside of the block. Talk about easy on and off! If you've dealt with prying a factory pan or a KillerBee pan off the bottom of a motor you know how they can be glued on there like they are welded on...and how easy it is to ruin the pan lip or bottom of the block if you stick a pry tool in there. With an O-Ring the pans comes off so easy. The problem is sourcing an O-Ring to make that work. Cosworth doesn't sell the replacment parts, at least not where I can find them. If you can figure out a way to use an O-Ring that would be a big advantage. The Cosworth has a very shallow groove for the O-Ring to sit in while you bolt it up.

Also on the ARE the front scavenge fitting is too close to the fitting where the DS tank returns to feed up thru the pan into the OEM oil pump. It makes it almost impossible to get an AN wrench on either hose. Leave enough room between them for a -12AN fitting + the AN wrench to turn it.

And finally leave room between the scavenge fittings and the motor mounts. On a STI it's very hard to get a scavenge line on the rear of the three available scavenge fitting ports. It seems like these pans are designed without taking the AN lines into account for being wrenched on there too.

[DanielsDM]

Zack,
Check out Nutter Racing oil pumps, reasonable prices and looks like a quality product. I was going to use one of theirs on my system, but I came across a good deal on a used Johnsons pump so I went that way instead.

I personally don't like the low pump mount setup on an 818. Just not enough airflow under the motor in a mid engine layout with a full bottom+diffuser. Just my $.02
I designed (and machined) my own pan so the fittings are at the front of the motor for better accessibility and keep the hoses away from the exhaust.
http://thefactoryfiveforum.com/showt...l=1#post288518

[Scargo]

Just reflecting on what Sgt. Gator mentioned from PRI 2017 and Nick Hooshow concerns about oil leaving the pickup points during acceleration. Daniel your pan looks like it could be prone to that.
Please explain what it is about the heat under the engine that bothers you. Air could easily be directed in there if needed with NACA ducts and the pumps wrapped or shielded. Custom exhaust work might be well worth it.

I really like the idea of an o-ring on the pan. Is Cosworth's all on the inside of the bolts? I believe McMaster-Carr sells o-ring stock for DIY lengths. Probably could be joined in a fashion like is done on headcovers with a dollop of silicone sealant in the splice spot.
Lastly, it seems many, or some, get by with just two scavenge pumps going to two scavenging spots on the pan. Seems if you have a front to back trough you could do just one port for the front and one at the rear. You'd then have acceleration and deceleration covered.

[DodgyTim]

https://www.ludowiciseals.com.au/o-r.../splicing-kits

It's quite common in industry to make up custom length o-rings using bulk length o- ring material and joining using cyanocrylate glue

Cheers

[Scargo]

https://www.ludowiciseals.com.au/o-r.../splicing-kits

It's quite common in industry to make up custom length o-rings using bulk length o- ring material and joining using cyanocrylate glue
Cheers

I've done that before myself, but didn't want to stick my neck out and recommend it. CA is sensitive to some things like moisture and chemicals (though there are many formulations). Viton is better and I'm not sure about glueing it together.
Thanks!

[DanielsDM]

Scargo,
The acceleration issue is something to consider. Since I read Gator's post I have given it more thought, may end up adding a pickup at the rear of the pan.
Regarding heat, I'm only ducting as much air as needed through the engine compartment, it will be hot down there. Also on top it's easier to access.
I considered doing an o-ring seal on the pan like Cosworth, and I still may go that way. It can be a PIA to keep the o-ring in the groove when installing though. Mcmaster sells custom length o-rings that are just cut to length and glued, they work fine for static seals like a pan.

[Zach34]

Zach if you make a Cosworth/Aviaid style pan I have a suggestion or two... I worked on a design but scrapped the project because I don't have a CNC machine and the cost to produce outsourced vs the size of the market just wasn't economically feasible. But here's a couple of thoughts for you:

The Cosworth pan uses an O-Ring instead of Three Bond or Permatex to seal it to the underside of the block. Talk about easy on and off! If you've dealt with prying a factory pan or a KillerBee pan off the bottom of a motor you know how they can be glued on there like they are welded on...and how easy it is to ruin the pan lip or bottom of the block if you stick a pry tool in there. With an O-Ring the pans comes off so easy. The problem is sourcing an O-Ring to make that work. Cosworth doesn't sell the replacment parts, at least not where I can find them. If you can figure out a way to use an O-Ring that would be a big advantage. The Cosworth has a very shallow groove for the O-Ring to sit in while you bolt it up.

Also on the ARE the front scavenge fitting is too close to the fitting where the DS tank returns to feed up thru the pan into the OEM oil pump. It makes it almost impossible to get an AN wrench on either hose. Leave enough room between them for a -12AN fitting + the AN wrench to turn it.

And finally leave room between the scavenge fittings and the motor mounts. On a STI it's very hard to get a scavenge line on the rear of the three available scavenge fitting ports. It seems like these pans are designed without taking the AN lines into account for being wrenched on there too.

Great point on the o-ring. As others have pointed out, it's a somewhat standard practice to cut and glue custom o-rings. I believe there are special adhesives. It's relatively low risk. If it leaked, it would just be a dribble when the engine is shut down. I plan on pulling a vacuum with the DS system.

Thanks for a heads-up on the ARE scavenge port locations. I'll do my best to factor that in on my prototype. It will just be a slow process...

[Zach34]

Zack,
Check out Nutter Racing oil pumps, reasonable prices and looks like a quality product. I was going to use one of theirs on my system, but I came across a good deal on a used Johnsons pump so I went that way instead.

I personally don't like the low pump mount setup on an 818. Just not enough airflow under the motor in a mid engine layout with a full bottom+diffuser. Just my $.02
I designed (and machined) my own pan so the fittings are at the front of the motor for better accessibility and keep the hoses away from the exhaust.
http://thefactoryfiveforum.com/showt...l=1#post288518

I can't believe I missed your post about your custom pan! That's awesome!

All the DS pans out there seem to be so different. I wonder where the oil falls onto the pan from the most. Hard to come up with my own design.

In the mean-time, here is the first piece. Adapter for the Peterson vacuum regulator.

20171219_233228.jpg

20171219_233302.jpg

[Sgt.Gator]

I can't believe I missed your post about your custom pan! That's awesome!

All the DS pans out there seem to be so different. I wonder where the oil falls onto the pan from the most. Hard to come up with my own design.

In the mean-time, here is the first piece. Adapter for the Peterson vacuum regulator.

20171219_233228.jpg

20171219_233302.jpg

Very nice. That's the way Cosworth does it too.

[Scargo]

I can't believe I missed your post about your custom pan! That's awesome!

All the DS pans out there seem to be so different. I wonder where the oil falls onto the pan from the most. Hard to come up with my own design....

I am sure someone could analize track data and figure out time spent at certain G values in lateral and longitudinal cornering.
From this data you would compare threshold cornering values to engine pumping speeds and see what kind of flow you have to deal with at certain times. With this analysed, shapes could be evaluated as to whether they can contain oil for the required period of time at high G's and high RPM's. Though it might be costly to manufacture, I wonder if the proper sized bulb shape might best contain oil as it drains down? This might apply to wet as well as dry sump pan designs.
hourglass-with-flowing-sand.png

[Zach34]

Well I'm sure the crankshaft is just slinging it all over the crankcase walls, then there's going to be the amount that is always draining from the heads. I think I like the Dailey/ARE designs

Obviously there is some disagreement...

Dailey:
Dailey.jpg

Cosworth:
Cosworth3.jpg

Aviaid:
Aviaid2.jpg

[Bob_n_Cincy]

I didn't use a vacuum regulator after someone had problems with seals.



One note: the vent line from the tank to the engine, the air can travel either direction depending on "blow by" and scavenge vacuum.

[RetroRacing]

WE did the same thing with the vacuum regulator, worked great on one engine, pushed the seals out of two others so we ditched it. I think you need at least 3 scavenge lines to pull enough vacuum at full boost to seal the engine and regulate vacuum in the block. I think we just had an engine with very little blowby and low boost that worked well with the Element system. We now vent to the tank.

We have our system plumbed exactly like you have above, but are changing it now to have the cooler and external filter between the scavenge pump and the DS tank, then another filter at the stock location. I am worried that we were losing too much pressure with that set up. WE are also moving the pressure sensor to the block and away from the external filter mount, regardless of how easy that was to wire.....

[Bob_n_Cincy]

We have our system plumbed exactly like you have above, but are changing it now to have the cooler and external filter between the scavenge pump and the DS tank, then another filter at the stock location. I am worried that we were losing too much pressure with that set up. WE are also moving the pressure sensor to the block and away from the external filter mount, regardless of how easy that was to wire.....

I definitely agree that the pressure and temp sensor would be better in an oil galley plug. Maybe the one near the PCV location which would be further from the pump. I also think having a remote oil pressure relief valve would be better at that location. We are not going big HP, so for 2018 we are going to run what we got now.

I'm not worried about loosing pressure as the big racing filter has much less restriction than the standard filter.
Our liquid/liquid cooler is made for a big block chevy (7.4L) that runs continuously at 6000 RPMs. We have the cooler plumbed into the cold water from the radiator.

Bob

[Zach34]

WE did the same thing with the vacuum regulator, worked great on one engine, pushed the seals out of two others so we ditched it. I think you need at least 3 scavenge lines to pull enough vacuum at full boost to seal the engine and regulate vacuum in the block. I think we just had an engine with very little blowby and low boost that worked well with the Element system. We now vent to the tank.

We have our system plumbed exactly like you have above, but are changing it now to have the cooler and external filter between the scavenge pump and the DS tank, then another filter at the stock location. I am worried that we were losing too much pressure with that set up. WE are also moving the pressure sensor to the block and away from the external filter mount, regardless of how easy that was to wire.....

Thanks for pointing that out. Probably saved me some big headaches. So, just to make sure I understand, your only crankcase vent goes to the atmosphere via the dry sump oil tank? You figure the scavenge pumps pull a vacuum through the vent at part throttle, but the flow reverses in the vent line to relieve crankcase pressure under boost?

[Scargo]

Yeah diminishing returns considered seems like a pretty well tested and thought out kit. I like the under mount for low GC and the reduction of hoses the RCM ZEN and Daily kits support. And it goes along with my thoery of K.I.S.S. And keeping it simplistic.

BTW they also sale the oil cover f/uw pump oil feed or filter feed: http://rogerclarkmotorsport.co.uk/st...ver-cover-feed. Along with some other bits and prices for doing ones own dry sump config.

BTW, you can modify a stock oil pump housing and return oil with this Moroso 23815 Oil Filter Block-Off Plate:
41t7Vx-JY-L.jpg

As to pressure buildup, why can't you have a pressure relief valve as well as a vacuum relief valve (and keep a closed system)?

[RetroRacing]

Thanks for pointing that out. Probably saved me some big headaches. So, just to make sure I understand, your only crankcase vent goes to the atmosphere via the dry sump oil tank? You figure the scavenge pumps pull a vacuum through the vent at part throttle, but the flow reverses in the vent line to relieve crankcase pressure under boost?

Yes. vacuum at low rpm is allot, so we felt pretty safe, but once the boost kicked in on long pulls, out went the rear main. We have all of the vent lines teed together and are venting through the oil fill with dash 12 to the oil tank.

[Sgt.Gator]

Retro were you running a larger pulley on the DS pump?

As I've pointed out before, the standard Element tuning size pulley, and the one that usually comes with the Aviaid pump, is smaller than the crank pulley, which means the DS pump is spinning at 130%+ of crank speed. In every pro system I've seen the DS pump ALWAYS runs on a larger pulley than whatever method is driving it, resulting 30 to 50% of crank speed. The Cosworth uses a SMALL drive mandrel and a large DS pump pulley, it drives at about 30% of crank speed.

I had no issues on my LGT with not enough vacuum I think because I used a larger pulley on the DS pump and an underdrive crank pulley that results in about 70% crank speed. That part is all detailed in this thread. My STI is running the Cosworth. I'll setup a guage to measure vacuum/positive crankcase pressure when the engine gets back from rebuild. (Coolant Failure, not Oil). The 818R will have the largest pulley I can fit on the Aviaid pump, 5.5" and either a full size Fluidamper or an undersized OBX crank pulley. Probably the undersized OBX because it's an EJ207 and likes to spin at higher RPM.

It's my contention that overdriving the DS pump at 130% crank speed results in two problems both related to the pump cavitating at high rpm, 1) heat and bubbles in the oil; and 2) lack of suction. The only way to really know is to put a combo vacuum/boost gauge on the crossover pipe and run fully sealed. If you are seeing positive pressure at all then go with the open system to the DS tank as in the drawing above.

[Bob_n_Cincy]

Retro were you running a larger pulley on the DS pump?

As I've pointed out before, the standard Element tuning size pulley, and the one that usually comes with the Aviaid pump, is smaller than the crank pulley, which means the DS pump is spinning at 130%+ of crank speed. In every pro system I've seen the DS pump ALWAYS runs on a larger pulley than whatever method is driving it, resulting 30 to 50% of crank speed. The Cosworth uses a SMALL drive mandrel and a large DS pump pulley, it drives at about 30% of crank speed.

I had no issues on my LGT with not enough vacuum I think because I used a larger pulley on the DS pump and an underdrive crank pulley that results in about 70% crank speed. That part is all detailed in this thread. My STI is running the Cosworth. I'll setup a guage to measure vacuum/positive crankcase pressure when the engine gets back from rebuild. (Coolant Failure, not Oil). The 818R will have the largest pulley I can fit on the Aviaid pump, 5.5" and either a full size Fluidamper or an undersized OBX crank pulley. Probably the undersized OBX because it's an EJ207 and likes to spin at higher RPM.

It's my contention that overdriving the DS pump at 130% crank speed results in two problems both related to the pump cavitating at high rpm, 1) heat and bubbles in the oil; and 2) lack of suction. The only way to really know is to put a combo vacuum/boost gauge on the crossover pipe and run fully sealed. If you are seeing positive pressure at all then go with the open system to the DS tank as in the drawing above.

Gator,
I talked to John at Aviaid about pump speed.
Since the pump is scavenge only. (air and oil mix)
There is no problem running the pump fast.
No problem with cavitation.
Bob

[Sgt.Gator]

John and I talked about it at PRI. I know he thinks that.
I still contend there is a reason Cosworth designed and manufactured a special crank bolt and drive mandrel so they could use a small drive pulley and a large pulley on the pump. Their's too is scavenge only. Since they also use the OEM A/c location and OEM a/c belt tensioning system with a few other Cosworth spec parts to make it work on the belt being further out, why didn't they just use the OEM a/c belt ring on the oem crank puley? You could argue that they wanted a cogged style belt, not a serpentine belt, and that might be the only reason. I think it's also to drive the scavenge pump at a much lower rpm.
All I know is what works for me. I'll be sure to have vacuum/positive crank case gauges, and eventually data logging, on both the STI and 818R this year. That's the only way to really know what is happening.

[Rob T]

I am running an Aviaid pump with the "standard" pulleys. They look about the same size. No doubt I am getting a lot of air our of the bottom of the motor. I am sending that straight through the Setrab cooler and into the spintric before returning to the tank. Last outing between 45 and 65 F, I did not get the oil above 150. I have a Setrab oil cooler that is ducted to the side scoops and has an intake in each one along with fans pulling air through the cooler. The fans are thermostatically controlled and probably did not come on. I need my car to be able to handle between 40F and 105 ambient if I am going to use it year round. John at Aviaid said the pump can handle 10,000rpm no problem. Gator brings up a good question about the amount of work/energy you put into the oil running faster than necessary. John told me the scavenge pumps flow 16gpm each at 3000rpm pump shaft speed and are capable of 275psi discharge pressure. I, too, use the tank to supply oil to the OEM pump and filter. The there is about 18" of head pressure to the inlet of the pump based on the location of the tank, the level of the tank and the location of the stock oil pump.

[RetroRacing]

I agree with Gator, we need to measure the vacuum at song, once the engine is broken in. I will see if we have some spare ports on our Haltech ECU to get some readings.

[Rob T]

I am back at the track Saturday, if it is not raining. Supposed to be 60F for a high. I disconnected one of the "forced" feeds to the rear mounted cooler. There were two of them, one from each scoop on the side in front of the rear wheels. So now, air from one leg will flow back to the cooler, but not pressurize it, as it can go out the other 3" opening. I am hoping this will raise the oil temp enough. Otherwise, I will disconnect the other "forced" air supply and let the cooler pull air from the engine compartment if needed. This will give me the flexibility of easily hooking up forced air or not.

As a note, I drained the air vent port reservoir from the dry sump tank. I had not done that in a while. I got out about an ounce of milky white liquid, which looked and felt to be mostly water in a little bit of oil, very well mixed. I am guessing the water vapor from the oil and a little bit of oil mist is going this direction. Just to be sure, I checked the oil and all looks good. I will do one more track session (the third) on this oil and then change it. I am hoping for higher temperatures with the oil this weekend. I'll report back.

[Mulry]

Rob, why are you running the cooler before the Spintric? IIRC, Gary Armstrong recommends running straight from scavenge out to the Spintric and then sending the non-air side of the Spintric out to the cooler on the way to the oil tank. Is it just packaging that makes that difficult for you, or is that a design decision for other reasons?

[C.Plavan]

Rob, why are you running the cooler before the Spintric? IIRC, Gary Armstrong recommends running straight from scavenge out to the Spintric and then sending the non-air side of the Spintric out to the cooler on the way to the oil tank. Is it just packaging that makes that difficult for you, or is that a design decision for other reasons?

Because it does not work that way. It bypasses the flow to the coolers and just dumps in the tank if it is before the cooler. Ask me how I know. Gary told me to run it after coolers after I called him out on it.

[Mulry]

Good to know, thanks Chad. I'm guessing that the pressure increase going through the coolers overwhelms the separation back in the Spintric?

[Rob T]

I am an Engineer, so this was fun for me. The spintric has no moving parts so uses the motion of the fluid through the passages for separation. Pressure drop was key. If you go through the spintric first, then to the cooler, the pressure drop of the "spun" (no air) oil going through the -12AN hose to the coolers and back to the tank is enough to change the split between the air side and the liquid side of the spintric. I know that some of the oil was going through the cooler this way, because the cooler got hot, but not enough of it. What was important to me was to know that the Setrab unit could handle air and oil, which it can. So now, 100% goes through the cooler first, then to the spintric and then to the tank. The difference was more than 50F oil temperature(lower). I now have to do less cooling to get the oil to temperature and that is why I disconnected one of the ducts from the scoops on the side and left the other port open to vent into the engine bay. This way, I won't be sucking in engine bay air, but cool air from outside the car, but I won't be forcing it through the coolers either. If the oil gets to 180F, the fans will kick on (they pull air through the cooler and exhaust right at the back of the car). I'll know more after Saturday.

[Sgt.Gator]

Good Info! Now the question becomes: Is the Spintric worth it or is it really required? It's a bit expensive for what you get!
I'll try to go without it for now. If I can't keep the heat down I'll add it later. It's probably cheaper than more AN hose, coolers, ducting, and fans.

[RetroRacing]

I can't wait to get the car back so that I can test some of this stuff. We ran ours for some really hard 30min test sessions and never saw oil temps above 190 the way we had it plumbed. Excited to see what it will do with the new plumbing system (cooler on the scavenge side rather than on the pressure side).

[Rob T]

I left the Spintric in for now. I only wanted to change one thing at a time. I also thought the bit of back pressure the spintric adds might aid in keeping the oil cooler more full. It certainly is directionally correct, if not significant.

[Zach34]

Retro, are you using a Spintric?

[RetroRacing]

Nope. There is a de-aerator in the tank which we think is enough, considering the massive amount of oil held in the tank compared to a stock pan.

[C.Plavan]

Would I use the Spintric again after using it, and seeing what it did? Short answer- Hell no. Never needed them on any other car race car I had. I was just desperate to find a fix when I was having all those oil temp problems and coughed up the cash.