I've been doing a little more testing on the dyno mule with my cylinder heads on it this week (up to 894 HP now). After the last go around a few weeks ago I had some conversations with a couple of my head customers, and John S (oldiron.fe on the forum) mentioned putting a window in the oil pan to see what was happening. Ever since I started with this engine on the dyno last September, I've noticed that at the higher RPM pulls I have a significant decrease in oil pressure, typically starting around 65 psi at 5000 RPM and dropping to about 50 psi at 7000 RPM, with the oil temperature around 180 degrees, using 10W-30 VR1 Valvoline oil. Higher oil temps of around 200 degrees make it a few psi lower, and I really wasn't too comfortable with those levels. I've seen this oil pressure issue before with vacuum pumps, so I figured that was a contributing factor, but I thought it would be interesting to see exactly what the oil in the pan was doing.
Since I was planning on pulling the pan to check the bearings, I decided to go ahead with the window, potentially wrecking the oil pan but hopefully learning a few things in the process. After checking all the bearings (which looked almost new), I cut a big rectangular window out of the Milodon 7 quart oil pan and bolted a piece of 3/8" thick clear polycarbonate to the side, using a bunch of The Right Stuff to seal it. The engine is using a Moroso windage tray between the block and the pan, and a Precision Oil Pumps high volume pump using the small spacer between the spring and the cup plug. I've also installed a new vacuum pump from Star Machine to replace the GZ Motorsports pump that I had been running previously. This pump gave me a few more inches of vacuum, and probably contributed to the latest HP numbers.
I set up a camera to take videos of the window in the oil pan while the engine was running, and during a dyno pull. I also rigged up a tach so that you could see the engine speed in the video, but unfortunately the tach I used didn't work, or else it didn't like the tach signal from the MS3-Pro. Regardless, I started testing the engine and then watching the videos. I was really surprised at what I saw. The first video link is below; this is with a total of 8 quarts of oil in the engine. It takes a while in the video to get to the actual dyno pull, which happens starting around 1 minute and 45 seconds in. The oil levels are marked on the pan in the video:
https://youtu.be/ZPL_05cKRKYWow, the amount of air entrained in the oil during the actual dyno pull is shocking! Lots of big bubbles circulating in the oil, certainly contributing to aeration of the oil coming out of the pump. As my dyno junkie friend Royce said, it looks like you are spraying a garden hose into a bucket of water. There appears to be about 6 quarts of oil in the pan during idle, and it drops to about 4 quarts during the pull. A while back on the Engine Masters show on Motor Trend on Demand, they did an experiment where they saw oil pressure drop as engine speed increased, attributed it to foaming of the oil, and solved the problem by taking some oil out of the pan. I thought I would try the same thing, so we pulled 2 quarts out of the pan, and ran the test again. Here is the video:
https://youtu.be/cQRGg-nI8m0Well, that didn't work, we still had the same drop in oil pressure as before up at 7000 RPM, we seemed to have the same amount, or even a bit more, of foaming and bubbles in the oil, and we didn't make any extra power. The oil level in the pan looked like 4-1/2 quarts during idle, and about 3 quarts at peak RPM.
The obvious next test was to remove the vacuum pump and see how it looked without the pump on the engine. We went back to 8 quarts of oil in the engine, and ran without the vacuum pump. Here is the video:
https://youtu.be/sd29Ye2E0Q4Unfortunately for this video I had decided to remove the tach (which wasn't working), and apparently after repositioning, the camera position was off a little compared to the last two videos, and we got some glare from the window that kind of obscured the oil. However, if you look closely you can see that the oil level in the pan drops quite a bit farther in this video than in the first one, and there still seems to be a whole bunch of foaming and air bubbles in the oil. BUT, the oil pressure stayed stable throughout the dyno pull. Unfortunately the oil pressure win was offset by a large HP loss.
Based on the fact that the engine has been behaving like this (oil pressure wise) for several dyno sessions, and that the bearings looked very good when I inspected them, I'm not overly concerned about the oil pressure issue when the vacuum pump is running. However, it seems like there ought to be some oiling system modifications that could mitigate this problem. One thing I tried with this new vacuum pump setup was a second vacuum port lower in the engine; this was recommended by Star Machine, and also Andy Miller at Olds Performance. I actually put a hole through the side of the block right across from the #3 main cap, and then milled a slot in the main cap to allow air from the crankcase into the slot, through the hole, and out to the vacuum pump. The hope was that since the hole in the block was shielded by the cap, it wouldn't draw in too much oil, but that didn't work, so I went back to a single line from the valve cover to the vacuum pump. I might try that again next time around with a baffle over the slot in the cap; Lykins also suggested going through the fuel pump opening in the timing cover, so I could try that too.
Another thing is the windage tray. The Moroso tray has louvers that strip oil off the crank, and several of those louvers point straight down at the oil level in the pan. You can see this in the videos when the engine is idling, one of those louvers is right above the window and oil is coming straight down. It seems like using the Ford windage tray, which has louvers only on the right side of the engine and shoots the oil horizontally through the louvers, rather than vertically, would be a better design. Also it seems like a kickout on the right side of the oil pan, designed to catch that oil and give it a chance to slow down before it hits the rest of the oil in the pan, would be good. It's also possible that a screen on the windage tray, like the Canton pan has, would lead to less foaming in the oil. Maybe no windage tray at all would work best?
Maybe the standard volume pump would yield an improvement. And of course if you ran a really, really deep pan, that also might help. The list goes on...
I'm sure there will be a million theories on the best approach to this issue, but the problem is there is very little data to back up the theories, and getting data requires changing up oil pans, windage trays, oil pumps, etc., which of course is a huge pain. But if I go forward with any modifications I will certainly post them here for general interest. I'm not sure if there is really HP to be had with a better oil pan setup, but an improvement in the stability of the oil pressure would be a worthwhile goal.