Reputable links for horsepower from windage tray/crank scraper

[SchmidtMotorWorks]

...
Just like UFO's the sightings decline after mobile phones with video cameras become common.

Meh.

Thank goodness: now when a fireball passes over there are usually hundreds of reports and multiple videos.

Windage is accepted science now. It is quite an amount of trouble to rig up an engine like this. As I mentioned, many reports from people that saw videos while in technical programs several decades ago. You yourself saw a film clip from the 1930s.

My high school had an automotive trade program 40 plus years ago. That's gone now. So is drivers ed.

Another example: 87 years ago the Ford V8 was new and exciting. Cutaway engines were interesting. That film also discusses the baffling that was needed in the oil pan to combat oil surging. The engine of choice for Clyde Barrow. [Engine below is a straight six for a truck]
https://www.youtube.com/watch?v=3RB3z1er9Sw
Screenshot_2019-05-31 1932 - The Invention of the Ford V8 Engine - YouTube.jpg

The video you recommended did not confirm your claims.

[houser45]

We have a wet sump road race customer who spent $1000 on a custom windage tray and crank scraper from one of the nationally known builders, plus another $500 for installation. Evidently he expected some quantifiable magic, but the dyno showed no horsepower increase.

We didn't promise any horsepower, but agreed on a road race engine with an hour of hard pulls, a windage tray/scraper should help keep the oil around the pickup.

Other than the Ishihara-Johnson site, where can we send him to read up on what to reasonably expect from a windage tray and crank scraper? Any links on here to increases on the dyno?

Most stuff on the internet is from people who are trying to sell you something. I don't know anybody who has the money to just test stuff and publish the results without having some sort of angle on it. I've done a fair amount of oil pan testing over the years for different projects and I've never found anything that consistently adds power other than a deep pan. The bigger the better for a wet sump system on the dyno. It would seem like scrappers and screens should help but I've never seen an improvement myself. I'd tell your customer to focus on a pan that works. That is, a pan that fits in the car, doesn't leak, holds pressure all around the course, and doesn't blow oil out the breathers. He'll find it challenging enough just to hit those targets.

This I totally agree with, my experience with this starts when i was a young man at 23 years old working in a machine shop building dirt circle track engines. My thoughts or beliefs may have zero dyno or race simulation testing of the oil system. But anyways i built a 383 sbc wet sump modified engine. It was the first one that we used an h beam rod instead of the smaller i beam 3/8 bolted scat or eagles. Limited by the cross member we had a standard style champ or kevco pan. With the correct amount of oil in the system it would start to foam the oil and lose pressure. We could of either took a quart out of the system or changed back to the i beams. We switched rods. This taught me that the best scenario to have is the deepest/largest oil pan i could use was very beneficial. I am sure other things could have been implemented to help like a scraper but in these type of builds you are so limited by space.

[Kevin Johnson]

...
This I totally agree with, my experience with this starts when i was a young man at 23 years old working in a machine shop building dirt circle track engines. My thoughts or beliefs may have zero dyno or race simulation testing of the oil system. But anyways i built a 383 sbc wet sump modified engine. It was the first one that we used an h beam rod instead of the smaller i beam 3/8 bolted scat or eagles. Limited by the cross member we had a standard style champ or kevco pan. With the correct amount of oil in the system it would start to foam the oil and lose pressure. We could of either took a quart out of the system or changed back to the i beams. We switched rods. This taught me that the best scenario to have is the deepest/largest oil pan i could use was very beneficial. I am sure other things could have been implemented to help like a scraper but in these type of builds you are so limited by space.

That is a good example of the morphology or surface contours of the rod being important in windage control. This is why it is insufficient to simply calculate the volume and surface area (for wetting) of the rod and leave it out of CFD analysis for flow.

[ptuomov]

The shape of a breathing hole is not trivial: ovate or rectangular (or...). Cross-sectional area is not a sufficient discriminant.

I looked at the figures in the paper and I now see why you're confused. The tested design are the actual tray designs that they've considered. They mostly resemble rectangular openings. The modeling is fairly accurate with about 9 million cells and model the candidate trays in a fair amount of detail.

What confused you, and I agree it's a little bit confusing presentation, is that one of the exhibits (figure 5) has elliptical red highlighting on top of the windage tray to indicate the location and area of the breathing holes. I believe those however are just highlighting, not part of the model. The model itself accurately reflects the shape of the holes, including the "flaps" covering the holes.

---

In any case, in practice, for gas flow simulations, minor variations in the shape of the hole are in my opinion unimportant. The location and cross-sectional area are first-order important. I want to make this point because simulations like these take currently maybe a day to run and dyno experiments of course even longer. I don't want to leave anybody under the impression that the exact shape of the holes makes much of a difference for gas flows. If you want to experiment with windage trays, it's my opinion that experimenting with anything else other than size and location of the holes is a waste of time as far as gas flows are concerned. To the extent that the shape of the holes and how they are covered make any difference they make difference to oil flows. That is, if the shape and covering makes a difference it makes a difference via the mechanism of changing how much oil rebounds to the crankshaft.

[Kevin Johnson]

...
The video you recommended did not confirm your claims.

I have seen a crankshaft spinning dry at 3500 (it is a blur), there is no way that the human eye could see what they claim at higher rpm through an oil coated window.

The video cited from 1937 directly refuted both of your claims in the above sentence. The window is also visible and the light is played on the interior of the crankcase (backlit or internally lighted).

You apparently have difficulty processing multiple information sources simultaneously. The phenomenological description of the mixed ligaments and droplet sizes conforms with the Ricardo Laboratories paper. Go back and reread at what rpm the crankcase was filled with oil mist which would allow the droplets to appear as ropes, taffy or tentacles.

[Kevin Johnson]

The shape of a breathing hole is not trivial: ovate or rectangular (or...). Cross-sectional area is not a sufficient discriminant.

I looked at the figures in the paper and I now see why you're confused. The tested design are the actual tray designs that they've considered. They mostly resemble rectangular openings. The modeling is fairly accurate with about 9 million cells and model the candidate trays in a fair amount of detail.

What confused you, and I agree it's a little bit confusing presentation, is that one of the exhibits (figure 5) has elliptical red highlighting on top of the windage tray to indicate the location and area of the breathing holes. I believe those however are just highlighting, not part of the model. The model itself accurately reflects the shape of the holes, including the "flaps" covering the holes.
...

You really do not want to go there.

The resolution of the SAE pdf images, in this case, is quite good. You can see the lead small openings in the tray which are roughly elongated isosceles triangles. These are not mentioned in the analysis (yet another problem). These are not the drainage holes mentioned which are represented on the perimeter of the tray as concavities.

---

In any case, in practice, for gas flow simulations, minor variations in the shape of the hole are in my opinion unimportant. The location and cross-sectional area are first-order important.

It remains an open question whether rectangular or ovate openings were used -- this is again the fault of the authors and not my confusion. The papers below illustrate the significance of the problem.

https://www.victoria.ac.nz/scps/about/s ... _pipes.pdf

http://citeseerx.ist.psu.edu/viewdoc/do ... 1&type=pdf

... I want to make this point because simulations like these take currently maybe a day to run and dyno experiments of course even longer.

Yes, dollar cost is likely the PRIME consideration.

... I don't want to leave anybody under the impression that the exact shape of the holes makes much of a difference for gas flows.

This is wrong.

... If you want to experiment with windage trays, it's my opinion that experimenting with anything else other than size and location of the holes is a waste of time as far as gas flows are concerned. To the extent that the shape of the holes and how they are covered make any difference it makes difference to oil flows -- that is, whether the shape and covering makes a difference to how much oil rebounds to the crankshaft.

Wrong, wrong, wrong.

This is why aperture size and crown is important in expanded metal sheeting (directional screening). The theses from MIT (and associated SAE publications) on the Duratec also discuss some of this.

To the extent that the shape of the holes and how they are covered make any difference [,] it makes difference to oil flows -- that is, whether the shape and covering makes a difference to how much oil rebounds to the crankshaft.

In the above charitable parsing of the sentence with an added comma, it is true that the shielding and various other factors can make a difference to rebounded oil.

Edit: I see you have edited the last paragraph after my capturing it in a quote.

[ptuomov]

Published papers are all limited by the length, so not everything can be explained. Nevertheless, that 2013 paper is in my opinion by a wide margin the most informative and complete paper on crankcase gas flows. In my opinion, most of the findings can be extrapolated to other large displacement cross-plane V8s with wet sump oiling and passive ventilation.

One thing I wonder about is this use of wire meshes or small "louver" holes. Car factories haven't by my understanding used wire meshes anywhere in the oil pan since the 1980s. To the extent they use louver openings, they seem to make them large -- almost never small. This seems to be the case even in cars and engines that employ other expensive solutions, so I don't think it's just the cost. Yet the aftermarket seems to love wire meshes and small louver holes. I am thinking that car factories with their design budgets have come to the conclusion that all the small-hole solutions are an inefficient way to use the space in the crankcase in terms of the tradeoff between crankcase gas pumping losses and preventing oil from rebounding back to the crankshaft.

[Kevin Johnson]

Published papers are all limited by the length, so not everything can be explained. Nevertheless, that 2013 paper is in my opinion by a wide margin the most informative and complete paper on crankcase gas flows. In my opinion, most of the findings can be extrapolated to other large displacement cross-plane V8s with wet sump oiling and passive ventilation.

It is an important paper but it is, by far, not an exhaustive treatment of the subject. An empirical counter-example to the connecting rod assumption was just presented (and unsolicited by me). Simply because information can be extrapolated does not mean it will be correctly extrapolated.

...
One thing I wonder about is this use of wire meshes or small "louver" holes. Car factories haven't by my understanding used wire meshes anywhere in the oil pan since the 1980s. To the extent they use louver openings, they seem to make them large -- almost never small. This seems to be the case even in cars and engines that employ other expensive solutions, so I don't think it's just the cost.

Car companies quite regularly use wire meshes in the oil pickup opening located in the sump -- like the modern big displacement cross plane V8 Hemi. I am not trying to be disingenuous -- recall that when Porsche and Triumph used wire mesh they also deleted wire mesh from their pickup heads. Some OEMs use perforated metal. OEMs have published warnings to be wary of poorly calculated open area and resulting restricted flow of aftermarket parts.

OEMs also have to worry about things like the black death and various other means of blocking small openings. The louver openings and the OEM scraper ribs in the VG30DETT tray and pan are fairly small. Notice that the scraper ribs are on the floor AND walls of the pan. I think you need to look at more engines.

...
Yet the aftermarket seems to love wire meshes and small louver holes. I am thinking that car factories with their design budgets have come to the conclusion that all the small-hole solutions are an inefficient way to use the space in the crankcase in terms of the tradeoff between crankcase gas pumping losses and preventing oil from rebounding back to the crankshaft.

I agree that the car factories are trying to save money with their design budgets and reduce the complexity of parts if at all possible. When I added internal screening to a factory CAD designed windage tray and a dedicated scraper (for a Factory Team Car) they picked up over 3.5% at peak rpm (their dyno testing).

[ptuomov]

Good point on the wire meshes on the pickups. If it's economical to use there, it would probably be economical to use elsewhere. Yet I still don't know of any factory windage trays made after 1980's that use wire mesh or small louvers.

[ptuomov]

When I added internal screening to a factory CAD designed windage tray and a dedicated scraper (for a Factory Team Car) they picked up over 3.5% at peak rpm (their dyno testing).

Do you have any documentation on that?

[ptuomov]

Published papers are all limited by the length, so not everything can be explained. Nevertheless, that 2013 paper is in my opinion by a wide margin the most informative and complete paper on crankcase gas flows. In my opinion, most of the findings can be extrapolated to other large displacement cross-plane V8s with wet sump oiling and passive ventilation.

It is an important paper but it is, by far, not an exhaustive treatment of the subject. An empirical counter-example to the connecting rod assumption was just presented (and unsolicited by me). Simply because information can be extrapolated does not mean it will be correctly extrapolated.

The volume and cross sectional area of connecting rods is so small that they don't make a difference to crankcase gas flows. However, they can make a difference to oil flows. I think that for large displacement V8s, one needs to understand those gas flows before one can graduate into thinking about oil flows.

[Kevin Johnson]

When I added internal screening to a factory CAD designed windage tray and a dedicated scraper (for a Factory Team Car) they picked up over 3.5% at peak rpm (their dyno testing).

Do you have any documentation on that?

I still have the engine they shipped to me.

[SchmidtMotorWorks]

...
The video you recommended did not confirm your claims.

I have seen a crankshaft spinning dry at 3500 (it is a blur), there is no way that the human eye could see what they claim at higher rpm through an oil coated window.

The video cited from 1937 directly refuted both of your claims in the above sentence. The window is also visible and the light is played on the interior of the crankcase (backlit or internally lighted).

You apparently have difficulty processing multiple information sources simultaneously. The phenomenological description of the mixed ligaments and droplet sizes conforms with the Ricardo Laboratories paper. Go back and reread at what rpm the crankcase was filled with oil mist which would allow the droplets to appear as ropes, taffy or tentacles.

You apparently have difficulty processing information honestly.

Sad that I have to state the obvious:
Without a strobe light it would be a blur, Smokey made no mention of a strobe light.
This is an attempt to mislead on your part.

Further, the video with the strobe light was made from an end view of the crankshaft.
The plastic windows windows Smokey wrote about were on the side of the oil pan.
This is an attempt to mislead on your part.

When you can present data that directly supports the claims about taffy on a crankshaft I will be interested to see it.
So far, all we have is your imagination.

[Kevin Johnson]

Published papers are all limited by the length, so not everything can be explained. Nevertheless, that 2013 paper is in my opinion by a wide margin the most informative and complete paper on crankcase gas flows. In my opinion, most of the findings can be extrapolated to other large displacement cross-plane V8s with wet sump oiling and passive ventilation.

It is an important paper but it is, by far, not an exhaustive treatment of the subject. An empirical counter-example to the connecting rod assumption was just presented (and unsolicited by me). Simply because information can be extrapolated does not mean it will be correctly extrapolated.

The volume and cross sectional area of connecting rods is so small that they don't make a difference to crankcase gas flows. However, they can make a difference to oil flows. I think that for large displacement V8s, one needs to understand those gas flows before one can graduate into thinking about oil flows.

I think you need to reread what he wrote. I am talking about gas flows. Put two H-beams side by side on the common journal in a big cross-plane V8. H-beams will capture and translate flow as well as block flow more than I-beams.

[ptuomov]

...
The video you recommended did not confirm your claims.

I have seen a crankshaft spinning dry at 3500 (it is a blur), there is no way that the human eye could see what they claim at higher rpm through an oil coated window.

The video cited from 1937 directly refuted both of your claims in the above sentence. The window is also visible and the light is played on the interior of the crankcase (backlit or internally lighted).

You apparently have difficulty processing multiple information sources simultaneously. The phenomenological description of the mixed ligaments and droplet sizes conforms with the Ricardo Laboratories paper. Go back and reread at what rpm the crankcase was filled with oil mist which would allow the droplets to appear as ropes, taffy or tentacles.

You apparently have difficulty processing information honestly.

Sad that I have to state the obvious:
Without a strobe light it would be a blur, Smokey made no mention of a strobe light.
This is an attempt to mislead on your part.

Further, the video with the strobe light was made from an end view of the crankshaft.
The plastic windows windows Smokey wrote about were on the side of the oil pan.
This is an attempt to mislead on your part.

When you can present data that directly supports the claims about taffy on a crankshaft I will be interested to see it.
So far, all we have is your imagination.

In my opinion, it’s not enough to have a strobe light and a window in the pan to learn anything useful about what’s going on inside a crankcase of a running engine. One needs mechanical shutters on the inside surface of the window and a synchronized camera. At high rpm, there’s only a minuscule time interval between the opening of the shutters and the window being covered by oil to the point that nothing useful can be observed thru it. You of course know this, but I just wanted to point out the high level of unintentional comedy in some of the claims.