Reputable links for horsepower from windage tray/crank scraper

[Kevin Johnson]

928 Motorsports uses the dyno testing data by Ed Peters on our 2.4 Dodge system as advertisement for their version of our 928 system; it's not a big deal to us. I am glad they trust the data that much.

[ptuomov]

928 Motorsports uses the dyno testing data by Ed Peters on our 2.4 Dodge system as advertisement for their version of our 928 system; it's not a big deal to us. I am glad they trust the data that much.

Whoever made up the numbers on that 928 MS page doesn't know enough about engines to even make up the data to be internally consistent.

I did email them this comment some years back n 2012, but it looks like it "didn't take":

In any case, here's my question to you. Your dyno data is here:
http://www.928motorsports.com/parts/ima ... tChart.jpg.
Perhaps I am missing something, but the torque and hp values are not
possible. Torque in ftlb = power in hp * (5250/rpm), d Torque in ftlb
= d power in hp * (5250/rpm) holding rpm constant. That is, both your
torque and power increase columns can't be true.

rpm reported hp change reported torque change hp implied torque change
4000 0 3 -
5000 2 7 1.90
6000 3 8 3.43
7000 5 10 6.67
7500 7 13 10.00

What's the explanation? Do you have the raw dyno run data, like the
dynojet files? If this is just an eyeballing error from separately
eyeballing the curves from a chart, I recommend you revise the table
to be consistent.

[Kevin Johnson]

[BILL-C]

Approx 12 years ago a road race team that we build engines for decided to install a different wet sump oil pan that would allow them to mount engine lower in the stock spec chassis for better handling. First race they had badly fluctuating oil pressure in certain turns and higher oil temp. They wisely skipped enduro and all extra practices and barely finished main event. At post race tear down, bearings looked TERRIBLE. The car handled better though, so they wanted to invest in some r+d on dyno to try and make this new pan work. We freshened the engine and installed on dyno with intentions of looking for where the oil was going. We installed valve covers with lexan windows and an inspection sight tube in intake manifold to see down into the lifter valley. We ran engine up to normal red line and noticed it was a little off on power and a dip in oil pressure at about 7k rpm that recovered a few hundred rpm higher. Adjusting oil level in both directions made problem worse. I put on every bit of safety equipment i had and stood next to engine and looked down into lifter valley and valve cover windows on next few tests. The oil was NOT collecting in the lifter valley or the cylinder heads. Because of the significant drop in oil pressure from 7000-7400 rpm, I can say that the oil wasn't sitting nicely on the bottom of the sump either. The fact that the pressure recovered at the end of the pulls means that it was not simply an aeration problem. Where was the oil? We spent the next week on the dyno trying to figure this out. We tested every different kind of pan we could get our hands on and even made our own with the features that proved to be effective. That super shallow pan ended up as an organ donor. When you have a poorly designed oil pan, i believe that the oil can be formed roughly into a hollow cocoon surrounding the spinning rotating assy. I don't think that the oil can fully engulf the crank like taffy in the augers of a taffy machine, but rather it dances around the od of the spinning parts, almost kept away from the oil flung off, but at a standoff caused by the escaping oil bouncing off of the inside of the crankcase and oil pan and trying to rebound to it's source. Oil falling downward from cam tunnel also contributes to issue. A big kickout at passenger side oil pan rail acts as a scraper to allow oil to escape and go down to a quiet, protected place in sump under windage tray and baffles.

[BILL-C]

The majority of our race customers run american v-8's, but we still build a couple dozen l-6 and 4 cyl race engines a year. Many import customers have supplied windage trays/scrapers from Kxxxx xxxxxxx and all have been rewarded with steadier oil pressure and longer bearing life. Never did A-B-A testing for HP gains though.

[Kevin Johnson]

928 Motorsports uses the dyno testing data by Ed Peters on our 2.4 Dodge system as advertisement for their version of our 928 system; it's not a big deal to us. I am glad they trust the data that much.

Whoever made up the numbers on that 928 MS page doesn't know enough about engines to even make up the data to be internally consistent.

I did email them this comment some years back n 2012, but it looks like it "didn't take":

In any case, here's my question to you. Your dyno data is here:
http://www.928motorsports.com/parts/ima ... tChart.jpg.
Perhaps I am missing something, but the torque and hp values are not
possible. Torque in ftlb = power in hp * (5250/rpm), d Torque in ftlb
= d power in hp * (5250/rpm) holding rpm constant. That is, both your
torque and power increase columns can't be true.

rpm reported hp change reported torque change hp implied torque change
4000 0 3 -
5000 2 7 1.90
6000 3 8 3.43
7000 5 10 6.67
7500 7 13 10.00

What's the explanation? Do you have the raw dyno run data, like the
dynojet files? If this is just an eyeballing error from separately
eyeballing the curves from a chart, I recommend you revise the table
to be consistent.

It's been a few years now but this question came up before. I believe that the hp numbers were obtained while running steady state (in order to get the oil and coolant temps stabilized at the indicated rpm). The torque values were obtained from rpm sweep tests. I am afraid that I do not know the rpm/sec rate. I do know that his dyno cell was air conditioned to control the humidity level and air temperature. They had to put off dyno testing when it was raining heavily because the system could not keep up with dehumidification. That is a bit different and more accurate than using correction factors. Chrysler Corporate trusted his numbers.

You could contact Ed Peters at Magnum Force Engineering located at 350 Gold Schaefer Rd, Fredericksburg, Texas, 78624 and ask him yourself. I suggest writing a letter to him as he rarely answers the phone.

Please be respectful and do not just show up and attempt to enter his property. In Texas and Florida, at least, that is a good way to get shot.

[ptuomov]

928 Motorsports uses the dyno testing data by Ed Peters on our 2.4 Dodge system as advertisement for their version of our 928 system; it's not a big deal to us. I am glad they trust the data that much.

Whoever made up the numbers on that 928 MS page doesn't know enough about engines to even make up the data to be internally consistent.

I did email them this comment some years back n 2012, but it looks like it "didn't take":

In any case, here's my question to you. Your dyno data is here:
http://www.928motorsports.com/parts/ima ... tChart.jpg.
Perhaps I am missing something, but the torque and hp values are not
possible. Torque in ftlb = power in hp * (5250/rpm), d Torque in ftlb
= d power in hp * (5250/rpm) holding rpm constant. That is, both your
torque and power increase columns can't be true.

rpm reported hp change reported torque change hp implied torque change
4000 0 3 -
5000 2 7 1.90
6000 3 8 3.43
7000 5 10 6.67
7500 7 13 10.00

What's the explanation? Do you have the raw dyno run data, like the
dynojet files? If this is just an eyeballing error from separately
eyeballing the curves from a chart, I recommend you revise the table
to be consistent.

It's been a few years now but this question came up before. I believe that the hp numbers were obtained while running steady state (in order to get the oil and coolant temps stabilized at the indicated rpm). The torque values were obtained from rpm sweep tests. I am afraid that I do not know the rpm/sec rate. I do know that his dyno cell was air conditioned to control the humidity level and air temperature. They had to put off dyno testing when it was raining heavily because the system could not keep up with dehumidification. That is a bit different and more accurate than using correction factors. Chrysler Corporate trusted his numbers.

You could contact Ed Peters at Magnum Force Engineering located at 350 Gold Schaefer Rd, Fredericksburg, Texas, 78624 and ask him yourself. I suggest writing a letter to him as he rarely answers the phone.

Please be respectful and do not just show up and attempt to enter his property. In Texas and Florida, at least, that is a good way to get shot.

I might also want to unhook my phone and shoot visitors if I had published torque and power numbers that don’t add up! .

(That’s just a joke, of course, I don’t know anything about the person or the test. Still, you probably understand some degree of skepticism about numbers in which torque and power don’t match up.)

[Kevin Johnson]

The majority of our race customers run american v-8's, but we still build a couple dozen l-6 and 4 cyl race engines a year. Many import customers have supplied windage trays/scrapers from Kxxxx xxxxxxx and all have been rewarded with steadier oil pressure and longer bearing life. Never did A-B-A testing for HP gains though.

I heard from that guy and he told me that he was pretty surprised at the interest from V8 owners because of the large existing parts aftermarket. Initially he thought that most enquiries would come from under-supported foreign engines. He was a Renault fanatic and had first hand experience in a paucity of parts being offered stateside. Archer Brothers Racing was an exception to this (they raced R8s on up through the R9/R11 (Alliance/Encore).

[Kevin Johnson]

...

I might also want to unhook my phone and shoot visitors if I had published torque and power numbers that don’t add up! .

(That’s just a joke, of course, I don’t know anything about the person or the test. Still, you probably understand some degree of skepticism about numbers in which torque and power don’t match up.)

He is very interesting to chat with. I believe he also ran one of Chrysler's plants down in Mexico.

[Kevin Johnson]

Modelling oil droplet/film interaction in an aero-engine
bearing chamber
M Farrall, S Hibberd, K Simmons
The University of Nottingham Technology Centre in Gas Turbine Transmission Systems,
University Park, Nottingham, NG7 2RD, UK.

https://www.yumpu.com/en/document/view/ ... ass-europe

The predicted trajectories are shown in Fig. 4 for oil droplets having diameters larger than
25 microns. It is observed that there is no appreciable difference in the oil droplet motion with
initial release location around the bearing cage. The large initial azimuthal component of the
velocity means that the oil droplets travel in a predominantly straight line until they impact
with the wall film on the chamber housing, where the majority have sufficient energy to
promote splashing. This results in several smaller droplets being ejected back into the core
flow, a significant proportion of which have diameters less than 25 microns. These droplets are
strongly affected by the core airflow and travel several circuits of the chamber before again
impinging on the wall, or exiting the chamber through the vent pipe with the air.

-snip-

5. Conclusions
A composite numerical model combining a commercial CFD code with sub-models for oil
droplet/film interaction and the motion of the oil film has been presented. Preliminary results
show that oil droplets shed from the bearing cage have sufficient energy on impact with the
chamber housing and any associated wall film to promote splashing. As a result, droplets
having smaller diameters (more likely to be affected by the rotating airflow) are ejected back
into the core flow. This increases both the residence time of the oil in the chamber and the
proportion of the oil carried by the sealing airflow through the vent pipe. The model provides
valuable information regarding the film thickness distribution around the chamber housing
along with the effect of geometric features such as vent and scavenge pipes that can be used
for future heat transfer studies.

Figure 2.jpg

Figure 3.jpg

Figure 4.jpg

[Kevin Johnson]

https://www.hotrod.com/articles/ctrp-06 ... dage-tech/

As one would expect, that testing shows a power loss from the windage tray in a stationary dyno:

D36C2439-6669-410A-A33D-F82BAE88A0DD.jpeg

This is of course not the same as lap times.

One could even make the argument that if you’re not losing at least a little bit of power on a stationary dyno from a windage tray, then you’re not doing something right...

The tray shown (below) has many smooth areas that would allow droplet impact and "splashing" with multiple smaller droplets ejected which would then result in a locally denser composite atmosphere of air and suspended oil droplets.

Windage tray design is complex and often must deal with diametrically opposed technical requirements.

ctrp_0603_pan_03_z.jpg

[ptuomov]

Oh so they just designed the tray wrong, so that’s why no gains on a stationary dyno?

What fraction of scraper and windage tray “dyno” gain claims on the internet are outright fabrications? 80%? 95%? 99+%?

This must be a very frustrating industry for an honest person. If it’s not enough to show only a small power loss on stationary dyno and better oil pressure on track to sell the product, all honest operators will be driven out of scraper/windage tray business.

[redgoldstein]

what is the best set up for a 440 mopar with a 7qt miladon center sump pan thanks

[Kevin Johnson]

Oh so they just designed the tray wrong, so that’s why no gains on a stationary dyno?

What fraction of scraper and windage tray “dyno” gain claims on the internet are outright fabrications? 80%? 95%? 99+%?

This must be a very frustrating industry for an honest person. If it’s not enough to show only a small power loss on stationary dyno and better oil pressure on track to sell the product, all honest operators will be driven out of scraper/windage tray business.

Hey, your friends at Chrysler have led you down a garden path. I just looked up the part numbers for the OEM "Baffle, crankshaft oil scraper" for the big cross plane V8 engines, namely the LA and RB. The part numbers are 2531945 (since 1968) and 2863983 (since 1967).

What percentage of what they are telling you is an outright fabrication? Are they good with Photoshop?



You had better investigate this carefully and document it.

~~~~~~~~~~~~~~~~~~~~~~

UPDATE !!!

The Chrysler developed windage tray is the best horsepower-for dollar item available for your engine, and a real industry first. It will fit all the 361, 383, 400, 413, 426 and 440 wedges with complete interchangeability. Basically, the windage tray provides increased horsepower by reducinq oil friction on the crankshaft assembly at high speeds. It’s worth an extra 15 HP at 6000 rpm! Remember, you must use two pan gaskets - one below and one above the tray!

[ptuomov]

Oh so they just designed the tray wrong, so that’s why no gains on a stationary dyno?

What fraction of scraper and windage tray “dyno” gain claims on the internet are outright fabrications? 80%? 95%? 99+%?

This must be a very frustrating industry for an honest person. If it’s not enough to show only a small power loss on stationary dyno and better oil pressure on track to sell the product, all honest operators will be driven out of scraper/windage tray business.

Hey, your friends at Chrysler have led you down a garden path. I just looked up the part numbers for the OEM "Baffle, crankshaft oil scraper" for the big cross plane V8 engines, namely the LA and RB. The part numbers are 2531945 (since 1968) and 2863983 (since 1967).

What percentage of what they are telling you is an outright fabrication? Are they good with Photoshop?



You had better investigate this carefully and document it.

Like the Tea leaf paradox? You think investigating that led to any other insights about crank scrapers and windage trays beyond identifying you as charlatan?

Do I understand this correctly that you’re trying to mislead readers to believe that there’s some credible, non-fraudulent dyno result buried in the library archives that will show that this factory part adds to power on a stationary level dyno?