Suddenly, it's 1953 again.

[PackardV8]

Few of you will find this of interest, but this is what the Vintage section is about; just tossing out the tribulations of being asked to help build a turbocharged Studebaker Champion 170" flathead 6-cylinder. It's now time to settle on a camshaft specification. Because few have done exactly this build, I've been digging through the files and running several DynoSim iterations.

Problem is, garbage in, garbage out. Because of the complications of putting a flathead six on a flow bench, we have to guesstimate the intake flow.

Over the years, I've found Studebaker's published horsepower ratings to be somewhat reliable, so I reverse calculated the CFM by juggling it until the results showed the published horsepower.

This brought me back to what Harley-Davidson Racing Division found with the KR engines. Flatheads want both flow and compression, but they can't have both at the same time. The most powerful mile track KRs would make one horsepower per cubic inch, but had looong valve timing and only 6.5 compression. They didn't make enough compression to be kick-started, but required a fast push start.

Back to the DynoSim - there are no hard data as to how changing the compression ratio of the Champion in the 6.5 to 8 range affects flow CFM, but it obviously must. If one only increases the input compression ratio without concomitantly decreasing the intake flow, the predicted horsepower increases, but that's bad science; the inverse relationship is inviolate.

Someone said, "Since it's a turbo, just increase the boost." Yes, No, Maybe. Back at the dawn of the aftermarket turbo, John Lingenfelter told me, "Every ICE wants to be turbocharged; the question is how much. Boost is the measure of work done without producing horsepower. The best design tradeoff is as much flow possible at the lowest possible boost."

Since this is a street engine, increasing the cam timing to KR levels is not desirable, but that's OK, because it's not physically possible either. Since this is a flathead, the cam/lifter interaction is akin to a bucket tappet OHC. I'm stirring around trying to find old Jaguar XK cam timing info, as those were long-stroke, low-revving critters, but again that cam timing data is not easy to find.

I'd hoped to be able to utilize a more modern, higher intensity profile, but when I brought one to my cam grinder, he said, "How many times I tole you; I can't grind air!" The Studebaker cam is a skinny stick (244 seat-to-seat) without enough meat on the lobes to add much lift, duration and intensity.

Any chance any of you have done a turbocharged flathead? All suggestions and questions gratefully appreciated.

[Porsche930]

Jaguar camshaft Specifications
Cam Lift (mm) 9.95mm
Duration 256° Inlet / Exhaust
Duration at 1mm 226°
Full Lift 111° Inlet ATDC / Exhaust BTDC
LTDC 1.29mm Inlet / Exhaust
Timing 17/59 59/17
Valve Lift (mm) 9.7mm
VC (mm) 0.3mm Inlet / 0.35mm Exhaust

[Tom68]

Super light intake springs, let the turbo blow them open.

Now about the exhaust timing, nah, I've got nothing, billet cam time.

Need to offset the cam and add rocker arms.

[Porsche930]

3.0 Porsche turbos run 6.5:1 comp. With a Garrett gtx turbo the pre boost rpm range is so small that the low compression isn’t a problem, it’s just not ideal for higher output levels. If you’re losing flow with higher compression, crank up the boost.

[Tom68]

3.0 Porsche turbos run 6.5:1 comp. With a Garrett gtx turbo the pre boost rpm range is so small that the low compression isn’t a problem, it’s just not ideal for higher output levels. If you’re losing flow with higher compression, crank up the boost.

It's not a comp problem, it's an adequate valve timing problem, exasperated by poor breathing.

[hoffman900]

Jack, as an aside, this guy has done some interesting modeling work and machining work on flat head cylinder head / chamber design: https://youtube.com/@wkuran?si=ha_UNn0o9P6_B7IF

Worth checking out

[Ken_Parkman]

I have that stuff on my bench. The difficulty is getting the shape of the port right for the 3D model. But it seems pretty good for assessing head differences and it makes a reasonable comparison to a more conventional OHV. Showed some surprises.

[PackardV8]

Thanks for the replies.

Wish I could get a Champion on your flow bench. From my limited experience, flow is the important half of it, but deck height/piston to head clearance is almost as important once the combustion process begins.

I hesitate to experiment with a customer's build, but this is a small bore (3.00") low speed <5,000 RPM, so I'm thinking one could go closer piston-to-head than one might try on big bore high RPM.

A ridiculous thought experiment, but it occurred to me that if the turbo only had to move dry air and with direct injection of alcohol, wouldn't that solve many of the flathead's volumetric efficiency limitations? Nah, answering my own question, it's probably not possible for a flathead to build sufficient compression to take advantage of alcohol, even turbocharged.

[Ken_Parkman]

It's reasonably do-able in todays age. The difficulty would be in the measuring of the block to make the model. Would you ever have some kind of factory block drawing? My dad sold his spare Champion engine.

If the bolt pattern was symmetrical (that would be too much to ask) you could cut a scrap iron head into 6 pieces, then use each one to do some different porting. Then you could start a database. Bill is making his own Kaiser head, so he wanted to do some more radical changes so he modelled and printed some different configurations.

The CFD is an interesting tool, but led to some wrong conclusions. After the flow test and some calibration it starts to make more sense. Of course the CFD is single phase which is clearly wrong, so you have to keep that in mind. But the more tools the more knowledge.

[PackardV8]

If the bolt pattern was symmetrical (that would be too much to ask) you could cut a scrap iron head into 6 pieces, then use each one to do some different porting.

Interesting you should suggest that, as it evidently also occurred to Studebaker Engineering. The Champion was built from 1939 to 1960 and over that twenty-one year span, there were six different head/combustion chamber designs. The last, 1958-'60, was the highest compression, 8:1.

Back to my question about piston-to-head clearance; how tight would you feel comfortable in this application?

[Ken_Parkman]

I have no experience with a flathead, so am not really a good person to comment. I can only say acceleration is proportional to the square of rpm, and linear with stroke and piston weight. Shorter rod length has an influence. All of those cannot be large numbers on that engine.

[jed]

jack google "turbo charging champion flathead 6" you should get Greg Myers pdf of his Studebaker champion 6 flathead build. Everything you need to know to complete the project is layed out with pictures and explenations. He runs at Bonneville and has set several records. There is a very short section at about 3/4th paragraph that states that flat heads due to small bore and long stroke run well with stock camshafts. Also another person Bill Cathcart is mentioned.

[PackardV8]

Hi, Jed,

Thanks for the suggestion. I know Greg and his build quite well. His solution of a two-layer CNC machined steel head is well-scienced. When I asked him how much to replicate it for me, "We did it for fun, but nobody has enough money to get another one built." So, not in the budget for a street build. I agree with most of what Greg has done and tested, but there are some other ideas I'd like to also try.

FWIW, I also knew Bill Cathcart well and he was a wealth of Champion expertise, but he's been dead for many years now.

jack vines

[FuelieNova]

No matter what a very interesting thread.
Keep us updated.
Tom

[Kevin Johnson]

It's reasonably do-able in todays age. The difficulty would be in the measuring of the block to make the model. Would you ever have some kind of factory block drawing? My dad sold his spare Champion engine.
...

I suspect if you contact the Studebaker National Museum that they would be able to make copies of the blueprints for an unknown price.