Please school me on dry sump systems

[Belgian1979]

I would like to know more about dry sump systems and although a lot of things are already on the net I haven't found a suitable answer for the following questions :
- what is the chance of a belt being thrown off ? I already had the accessory drive belts being thrown off, so this is at least something I would want to know about.
- With the tank being above the level of the oil pump, doesn't the dry sump tank drain it's oil to the pan when not in use ? How do you deal with that.
- Is it required to prepressurize a dry sump pump with an electric drill or is this because the pan needs to be emptied of the oil that drained into it.

Thanks

[Barbapapa]

I would like to know more about dry sump systems and although a lot of things are already on the net I haven't found a suitable answer for the following questions :
- what is the chance of a belt being thrown off ? I already had the accessory drive belts being thrown off, so this is at least something I would want to know about.
- With the tank being above the level of the oil pump, doesn't the dry sump tank drain it's oil to the pan when not in use ? How do you deal with that.
- Is it required to prepressurize a dry sump pump with an electric drill or is this because the pan needs to be emptied of the oil that drained into it.

Thanks

Those are good questions.
The reservoir will possibly drain back into the pan over a weekend depending on where your pump is mounted height wise. I've dealt with that and it's been discussed here by others.

Some dry sump pans are made shallow to mimic race cars that have entire drivetrain lower by using smaller flywheels and bellhousings.

In reality, most non-pro racers are not going to have that level of effort and there's no need for a pan to be unusually shallow and above the minimum ground clearance. There's no benefit from less volume.
I changed my own dry sump pan to use the full depth of a stock pan and am happy. Now much the oil can drain back into the pan over time but gets sucked right back out within seconds of starting. The deeper pan keeps the seconds after startup from being an oil immersed mess that will generate a puff of smoke as it gets by the rings.

Those with higher mounted pumps will never deal with drainback. IMO it's not helpful to install some kind of shutoff valve on the feed line which will probably be contemplated.

The key to belts is to have the "walls" only on one pulley. That will track true as long as everything is aligned and rigid. Good practices such as beefy mounting and laying straitedges on pulleys to ensure alignment is all that's needed. The timing belt systems don't require a lot of tension and track well, nothing like the archaic Vee-belts you struggle with.

Most pumps have a pulley clamping method to move the pulley anywhere along the shaft and fine tune alignment. Some also have a drive hex so you can prime.

[mbrooks]

The belt systems used are pretty reliable. Mine has been sitting for quite a while now and the tank hasn't drained a noticeable amount. If you route your pressure line low so the oil doesn't drain I don't see any need to prime an engine that has already been ran, and if you do feel the need to prime, there are pulleys available to hook to a drill using a longish belt.

[Hardcharger55]

The hose out of the tank on the bottom goes to the pressure side of the pump. It's possible for the oil to get past the pump, the filter and the oil cooler but its not very fast or very much. The scavenge side of the pump goes to the top of the tank.
Basically, your tank will not drain to the oil pan.

[robert1]

My guys run a lot thicker oil than you do. In a weeks time some of them will drain the tank in to the pan. If the pan is lower than the bottom of the tank unless you've figured out some way to defy gravity it's going to drain. Belt MUST be run lose.

[Belgian1979]

Ok, then a couple of other qestions :
- is a dry sump specific pan necessary ? Or can a conventional pan be converted by adding scavenging lines. I'm not thinking about an ultimate type of setup but I want mainly reliability and non aerated oil.
- If you route the line so that it makes a bend (a U if you will with the bent of the U higher than the oil level, this would prevent the draining. I'm not sure if it would cause too much of a retriction...
- I read somewhere that there were in pan dry sump pumps made (Barnes) Are these still available ?
- how close does the tank need to be to the pump ? I need to mount it in the right fender at the rear of the engine. I was thinking on putting the pump on the passenger side.

[andyf]

I've used deep wet sump pans as dry sump pans before. It works okay but it might not ideal. Just depends on the application.

[robert1]

It's been years since I built a motor with one of those Barnes setups. I had actualy forgtotten about those. They use a special pan. On to the wetsump pan for a dry sump yes you can weld bungs in and use an external pump. As far as a bend in the line no if the pan is lower than the tank no matter what you do to the lines as far as bends or raising then up it will not change the syphoning effect.

[MadBill]

High G acceleration makes it harder for the suction side of the pressure stage to draw oil from a rear mounted tank, but I've worked with cars having as low as 3 lb./HP and tanks behind the right rear wheel that worked just fine. I wouldn't try it in a fast drag car though, due to the potential for cavitation..

Also, a high-mounted tank will most certainly drain back into the pan through the pressure stage; much faster if the pump and internal clearances are on the high side. Johnson pumps are claimed to be very good in this regard.

[Cubic_Cleveland]

Belgian, I think the answers to your oil system issues are easier to find than engineering a dry sump system, and a LOT cheaper. Either way the pan has to come off and the oil pump come out so why not see what you can find first?

Aside from that I am a dry sump fan, but you have to be careful about trash getting into your belt system. Some sort of belt cover, or an enclosed/protected engine bay go along way to prolonging belt service life. It's not pretty when a belt lets go at RPM.

[Belgian1979]

Belgian, I think the answers to your oil system issues are easier to find than engineering a dry sump system, and a LOT cheaper. Either way the pan has to come off and the oil pump come out so why not see what you can find first?

Aside from that I am a dry sump fan, but you have to be careful about trash getting into your belt system. Some sort of belt cover, or an enclosed/protected engine bay go along way to prolonging belt service life. It's not pretty when a belt lets go at RPM.

Yes, I realize the belt is a failure prone item in the whole system and maybe even the biggest issue.

Bit of topic here but FWIW : I've been talking to several people about the oil pump issues. So far most are leaning towards the bypass being the culprit. The best answer so far came from Schumann pumps. I asked them what they suggested and basically they suggested a pump which bypasses to the sump instead of to the inlet. Their reasoning is that in a high bypass situation such as in high rpm, the bypass oil stream in the inlet backs up towards the pickup in essence restricting the inlet, inducing cavitation, which ends in a sort of oscillation at the bypass valve.

I've had other people that said basically the same thing when they analyzed the pump graph. Canton was not able to help me, other than they said that none of the billet pumps (which are very expensive) fit the pan without reconfiguring the rear wall. The billet pumps are better in that they have more inlet area.
The pump I use now in essence has a capacity to produce around 50 gpm at 6000 rpm. The volume of oil it produces certainly is not the problem.

I made a pressure drop calculation and if these calculations are correct, and the limit of 3-4psi is correct, I should still be ok with a pressure drop of around 2,4 psi at 6000. It's close but it might work. I'm inclined to try the external bypass first, maybe even combined with a metered hole in order not to put the internal bypass out of order entirely. If it doesn't work out, it will probably end up with a dry sump.

Like you said I need to protect the drive belt somehow. This is something not easily done, as I already would have to notch the front crossmember to have the pulley clear the crossemember, let alone a protective shield.

That being said, I hate experimenting with something as crucial as an oiling system, but I fear that other than restricting the engine to sub 5500 rpm use I do not have a lot of choice. That would defeat the whole idea of the engine though, since it would have been a whole lot easier to build a sub 5500 rpm engine.

[lorax]

A pressure relief that dumps oil back into the pan does have its advantages, and does help with high speed cavitation. System 1 sells a pump that dumps the oil directly back into the pan.

HOWEVER, you still seem convinced that some how a relief activated by pressure, set to open at 70+ psi is opening at 55-60. I don't understand your reasoning.

How ever, as has been posted on this site recently, and on others, there is a chance that the relief is not closing all the way and is bleeding of pressure, but is closed enough to have pressure be 70+ psi cold. Its happened more then once.

[Belgian1979]

While keeping an eye on the pressure gauge last time, it went to 75 at around 3000 and untill about 4000 after that I start seeing the pressure drop with 60 at around 5000 and going to 55 psi at around 5500. The pressure drop of was more pronounced and faster with the higher wt oil (although I didn't check at which point it reached 75). So it seems that it goes on the bypass but earlier.

It's a PD pump, which should normally increase its volume the more rpm its turning (if it can draw in the oil of course). Imo the evidence is it produces enough volume because it reaches the bypass setting at 3000.

If I'm not right, I should see the pressure drop becoming apparent at a lower rpm agreed because basically the inlet would have to draw in the total pumped volume. If that is the case it would only leave the pickup/inlet or the total displace volume at question.

[Kevin Johnson]

Bit of topic here but FWIW : I've been talking to several people about the oil pump issues. So far most are leaning towards the bypass being the culprit. The best answer so far came from Schumann pumps. I asked them what they suggested and basically they suggested a pump which bypasses to the sump instead of to the inlet. Their reasoning is that in a high bypass situation such as in high rpm, the bypass oil stream in the inlet backs up towards the pickup in essence restricting the inlet, inducing cavitation, which ends in a sort of oscillation at the bypass valve.

Be sure to read SAE 750051 from Mercedes where they show that a bypass increased sump oil aeration. Be careful where you direct your bypassed oil. Also remember that internal bypass recirculation helps give any entrained air bubbles additional time to dissolve into solution because of the higher pressure plus it helps decrease the average bubble diameter which also helps the air to dissolve.

I've had other people that said basically the same thing when they analyzed the pump graph. Canton was not able to help me, other than they said that none of the billet pumps (which are very expensive) fit the pan without reconfiguring the rear wall. The billet pumps are better in that they have more inlet area.
The pump I use now in essence has a capacity to produce around 50 gpm at 6000 rpm. The volume of oil it produces certainly is not the problem.

Hmmm. When I looked at the graph I thought it was basically saying the pump itself is not the issue because they have a more restrictive pressure side orifice than is present in your system (bypass kicks in earlier than on your engine). Maybe people are thinking that the bypass is defective somehow. My estimate for the pump is slightly lower (corrected for engine rpm versus pump rpm) of 20 to 22 gpm at 6000 engine rpm.

I made a pressure drop calculation and if these calculations are correct, and the limit of 3-4psi is correct, I should still be ok with a pressure drop of around 2,4 psi at 6000. It's close but it might work. I'm inclined to try the external bypass first, maybe even combined with a metered hole in order not to put the internal bypass out of order entirely. If it doesn't work out, it will probably end up with a dry sump.

That will be an interesting experiment. The hypothesis that I put forward of existing air entrainment in the sump oil via windage makes the prediction that the external bypass will worsen the situation (via splashing of the ejected oil into the reservoir and decreased ability to drive entrained air into solution). Try to make your modifications reversible just in case the hypothesis is not rejected.

[Belgian1979]

Kevin,

It was explained to me that the linear line before 2000 engine rpm is the volume displacement of the pump, above 2000 rpm in that graph the bypass seems to kick in (graph going flatter). If you take the gpm at say 600 engine rpm and you multiply by 10 for 6000 rpm, it would be a fair assesment, since the teeth displace the same volume but they turn faster (all other variables ignoring). However, even with 22 gpm the pump output should be sufficient against the backpressure in the system to reach the bypass setting. I cannot believe the engine would warrant 22 gpm not even on high rpm.

I agree on air entrainment. It will be fully reversible in the sense, that I just have to pull the exit line on the block filter mount off (-10) put a F/F connector in, hook up the external bypass and connect the bypass line to the fitting on my pan which is normally meant for the temp sensor (1/2 npt). The additional advantage is that the area where the temp fitting is, is behind a trap door.

No problems, unhooking it and reconnecting as Original.

We're getting a bit off topic, but there is a possibility that this would not provide the solution. If that's the case I think the only viable option would be dry sump. Right ?