Performance tuning the Stag engine

[richard w]

Complete thread drift, but congratulations on winning the V8 autosolo Neil.

[flying farmer]

John Eales always produced good stuff.

I had one of his early 3.9 V8 motors, before Rover started producing them. Generally he understated the power outputs which meant everyone was happy except perhaps his competitors who always seemed to underperform on group tests!

Another graph showing comparisons of exhaust manifolds. All cars have the same straight though systems and silencers
The black line and the blue line are the same engine in different cars. The black line is the engine currently installed in the Estate with Stag type 34mm exhaust manifolds, complete with original Stromberg carbs.
The blue line is from when it was installed in the TR with 38mm 4 into 2 into 1 manifolds.
As can be seen there is very little difference between the two except in the sub 2000rpm area. Top end power was within a couple of horsepower. The limiting factor here is the inlet manifold. When I fitted efi to the Stag manifold I couldn't feel any difference in the power output but I gained a little fuel economy.
The red line comes from my P reg Stag and was the result from this weekend. This is a 4 into 2 into 1 design the same as the one on the TR, but uses Stag size 34mm primaries and 38 mm secondaries. The TR manifold used 38mm throughout.
Despite the EFi and individual inlet runners it doesn't make the torque. The inlet manifold is a direct copy of the one on the TR It didn't make the torque with the fast road cams either. Driving the car it feels sluggish due to the lack of torque, and this engine has a 10:1 CR whereas the other only has 8.5:1 CR.

Maybe it needs the balance pipe moving further away from the engine. I did this on my Rover V8 engine Toledo last year and it transformed the torque delivery.

Something will have to be done as I am not happy with it at the moment. Yet another project for next Winter as well as swapping it to Megasquirt.
standard cam torque curve.jpg

Next time I shall do comparative graphs of when the fast road cams are fitted. This is when it starts getting interesting, and not without surprise and disappointment (as usual)!

Neil

There is definitely not a "one size works wonders for all" type of manifold, and I reckon there are at least three depending on the final target output of the engine.
Neil

Oh yes there is.
The best size is going to be related to the bore size and valve area.
The larger you go the less bottom end you get, and the gains at the top are suprisingly small.

For a 32mm valve with those (p..ss poor) flow figures, you don't need to be exceeding 35mm pipe sizes cos the exhaust valve is hardly even breathing close to the asthmatic Spitfire!

With exhaust port flow exceeding 60-65cfm you can be up at 38mm.

Have you seen the JE range rover V8 manifolds.
They are a lesson in how to do it.

JE-Big-Bore-Manifolds-for-Rover-V8.jpg

The rest is very dependent on camshaft duration and firing order, which we all know very well is a problem on all classic firing V8s.
Then quite simply there is NO ROOM in the engine bay of the Stag.

The way you get around that is making completely assymetric camshaft timing, which is exactly what they found to work brilliantly on A series minis.

[Tdm]

If only Triumph had spent the time and money on the Stag engine that you have Neil, we'd all have been a lot better off!
Fine work as usual

Yes quite agree - the car could have been turned into a real World beater if only it had been developed and managed properly.
All the ingredients were there - great looking timeless design, a great sounding British V8 engine, 4 seats with a perfect driving position attainable, hard & soft top option for driving in all weathers, and a car that made you feel good, both before, during, and after driving it.

[mike@thenook]

If only Triumph had spent the time and money on the Stag engine that you have Neil, we'd all have been a lot better off!

Fine work as usual

[TR5convalescent]

Thanks for the update, Neil.

[flying farmer]

Time to dig up this thread again!
I finally got round to trying my P reg Stag on the rollers again at the TR register event at Malvern yesterday. Apparently I have used North London Dyno 13 times now over the years! No wonder they know me by name (as well as the mad farmer) Quite a few other rolling roads over the years as well.

This Stag was a cocked up attempt to copy the engine spec of my TR. It failed miserably never getting better than 198bhp against the TRs 216bhp, and after further fiddling ended up with 190bhp but a much better torque curve after altering the inlet trumpet lengths a little.

The only difference was the exhaust manifolds. On the TR they were the same 4 into 2 into 1 design but I used one and a half inch tubing throughout their length. On the Stag manifolds I used the normal inch and three eights (or thereabouts, I cut up a set of regular Stag tubular manifolds to make them), for the primary pipes before going to inch and a half for the secondaries.

I came to the conclusion I was wasting my time with the fast road cams with these manifolds. When I built my K reg Stag I swapped the fast road cams for standard ones. It didn't feel much different, only slightly slower, and after several years I now have some results to confirm it.
It actually did better than I thought because it feels like a right slug compared to my other stag, although it is always dragging the extra weight of a hard top with it.

I have added the last set of figures for the fast road cams to the graph below. I have lost about 11bhp, but to be honest the fueling was dreadful, far too rich at the top end of the rev range with lots of black smoke. This car utilises a standard Rover V8 flapper type injection system. The over rich problem is easily got around in normal driving by not fully opening the throttle. Even at max rpm this engine is producing 70bhp less than the size of the throttle disc is capable of, therefore only using half to three quarter throttle is fine for most occasions. Next Winter it will be fitted with the Megasquirt ecu that the TR is currently fitted with and the problem will be fixed.

Standard and fast road cams compared.jpg
On the picture below I have shown the two torque curves (power figure is the wheels figure)

Note the two different air fuel ratios. The fast road cams start too rich and end up about right, the standard cams start about right and end up horribly rich, probably because the standard cams are pulling less air but the amount of fuel supplied stays about the same. The only adjustment for the fueling is the spring tension in the air measuring flap. I generally adjust this for the best cruising mixture, but the whole set up works better on an engine that is pulling more air. It was very close to requirements on the 216bhp TR engine.
I have taken the liberty of inserting Stagdads last set of rolling road figures into the graph (the red line). As can be seen it is pretty much smack in the middle of the two lines.

I suspect that if the mixture was corrected Steve's figures would be slightly better at the bottom end, and mine would be slightly better at the top end. Of course, by the time I get my next set of figures he will have built his new inlet manifold and the goal posts will have moved!
small bore manifolds compared.jpg
Over the next day or two I shall post a few more comparative graphs of the info I have obtained over the years with different bore exhaust manifolds. I now have a pretty good idea of what works, what doesn't, and what hopefully will in the future even if it doesn't now! God only knows how much money I have spent on stainless tubing over the years and the hundreds of hours welding together "bunch of bananas" exhaust manifolds. There is definitely not a "one size works wonders for all" type of manifold, and I reckon there are at least three depending on the final target output of the engine.

Watch this space.......

Neil

[KOY 23]

Good point about the tyres! Thanks.

[flying farmer]

Hi Neil.
I know that certain Dyno makers will only state wheel bhp, as calculation back to flywheel is affected by different losses in manual and auto boxes, which makes bhp comparisons impossible according to Dyno Developments.
Mine is a ZF, if the test was in 4th the torque converter would have locked up at 2000rpm.
Dont know if this helps.

Most manual gearbox dyno runs are done in 3rd gear, there are more transmission losses than in 4th as the power has to pass through the laygear rather than straight down the mainshaft. On the other hand there are less tyre losses, but I suppose the software takes account of all this.

I think wheel speed is the main reason they don't use top gear. I have heard that depending on diff ratio the gearing can be as high as 30mph/1000 rpm with the ZF in top gear. I think taking the tyres to 180mph in top at 6000rpm will probably result in a bit of a mess!

At the wheels figures are the only ones that can really be relied upon, I would have to agree with Dyno Develpoements on that point.

Neil

[Carboy0]

I started with the TPS - good tip. And then I read the rest. I agree with the point about fuel drop out due to low air speed etc.

[KOY 23]

Hi Neil.
I know that certain Dyno makers will only state wheel bhp, as calculation back to flywheel is affected by different losses in manual and auto boxes, which makes bhp comparisons impossible according to Dyno Developments.
Mine is a ZF, if the test was in 4th the torque converter would have locked up at 2000rpm.
Dont know if this helps.

[flying farmer]

Different Diff ratios must effect the readings too don't they Neil?

I think this is why the rpm is taken from the rollers, it will also take account of different tyre size so the computer can calculate the overall ratio all the way from the flywheel to the road.
I did have to ask the rolling road operator about this as I wondered how it all worked!

This is where the torque converter slip problem comes into the calculation.

If you watch how the rolling road operators set things up, they run the car up to roughly half maximum revs in the gear they will be using for the run, then hit a button and this will be the ratio used for the calculations.
I am having to guess here, but since this is done at a light throttle setting there is going to be more torque converter slip on an auto when full throttle is applied.

Also I believe a torque converter is multiplying torque while it is slipping but the dyno will only see the road speed so I would assume it would under estimate engine RPM, but the torque measured will be the actual albeit multiplied figure. I would imagine this is less of a problem at high rpm.

I do wonder if this is why Johns peak torque is coming in at only 3100rpm.

I know some gearboxes lock the converter at certain speeds/revs, but I don't know enough about it to know if this can apply on the rollers.

If anyone does know the answer to these questions please enlighten me!

Neil

[Stagdad]

The Guy that did my run also had to use the tuner studio rpm read out as I haven't figured out how to wire the dash Rev counter yet.

[Bob Heritage]

Different Diff ratios must effect the readings too don't they Neil?

[flying farmer]

That's why I wanted to find a dyno dynamics roller Bob as they have a shoot out mode which is supposed to be pretty consistent machine to machine. I'm sure Neil can advise if this is the case.

There is a lot of variation between rolling roads, but the dyno dynamics machine are the ones that usually appear at the Stag and TR national days, and was also the type that I got my 246bhp figure from.
In theory the software should make allowances for temperature and air pressure, but different makes of machine will no doubt use different ways of calculating the flywheel figures which is where the biggest variation could come from.
Having done lots of rolling road runs over the years, I have found the dyno dynamics machines to be very consistent between different operators, but I dare say there is equal consistency between other makes. It is just more likely that my cars will be run on a dyno dynamics machine as they are portable so are more likely to appear at places I am going!
Actually one of the biggest discrepancies that can appear is due to inaccurate rev counters. The RPM is usually synchronised to the rollers at about half maximum revs, and the calculations throughout the rev range are done from this. If the rev counter reads incorrectly then the peaks will occur at the wrong rpm.
I am told it doesn't actually alter the figures, just the rpm they appear at.
When I had my 246bhp stag on the rollers we used the figures taken direct from the display on the laptop, and it took him several attempts to get it synchronised to his satisfaction.
The rev counter on my TR is more of a guestimate at most revs, but I try and get the megasquirt display and the rev counter to agree at 3000rpm. Since it goes into the red at 5500rpm and the engine revs to 6800 I am glad it has a built in rev limiter!

Auto boxes are a real PITA for dyno operators as torque converter slip at low rpm can mess up the figures.

Neil