Flat HP Curve

Chassis dyno run.

The torque curve looks like a common one, but the flat HP curve has me puzzled.

Anyone have some input on why it’s this flat?

The huge drop off after peak HP is when he got out of the throttle. Around 6000.

Doesn't this normally indicate an engine that is "air starved"?

Meaning, the HP numbers aren't being produced because the motor isn't able to either pull in the air needed to make that power, or perhaps is mechanically starting to run into stuff like valve float (although in that case the curve just drops right off, so I don't think that's the situation you're seeing).

Regardless, I am very much in the "read about it" category here, as opposed to "been on the dyno and here is why it did that", so I'll sit back and enjoy the more experienced folks provide the real hints and/or answers.

Converter slip.

Would look like a "normal" graph with a locked converter or manual trans.

Same as my 440 showed , imo its the converter stall point . Seemed more noticable the bigger the stall,
I just made sure afr is right

Tex

Real unusual for peak torque and peak HP to occur at the same RPM other than at 5250 RPM
Was it spinning the tires at all or do you know ?

Chassis dyno graphs with an automatic transmission can often be very weird. Pull the engine and put it on an engine dyno and you'll see a totally different curve.

Rev limiter? A friend couldn’t figure why he could run different ETs but always ran 112 mph (8th). Finally after about 6 runs he saw it was hitting at about 75’ out from the finish line. He was running 6.10 to 6.15s in a 64 Plym

How would torque decay at the exact same rate as RPM increases, resulting in the flat line?

First time for me to witness a chassis dyno test.

I have more questions than answers. First question I had was rpm.
I asked the operator about engine speed on the dyno readout, he said it came from dyno roller speed but was only accurate if the trans was 1 to 1. It was in high gear the whole run, and the tach in the car was pretty close to the engine speed shown on the monitor.

We picked up 20 hp by messing with timing. Only change made.
Not my car, was just an invited bystander.

horsepower is a calculated value, The dyno measures torque, the engine rpm is then used in the calculation Horsepower (HP) = (Torque × Speed)/5252, if the dyno computer does not have the real engine rpm the readings are not correct. I used a Mustang dyno for several years, they either were too lazy or did not have the proper equipment to read engine rpm. The old roller rpm trick is when you can not get a real engine rpm number. Mustang used to use what was called smart tach, and it was sometimes a challenge to get a proper rpm reading, as a regular distributor, MSD, coil on plug, waste spark, etc. all have different settings in the soft ware to condition the signal. If the vehicle has a manual transmission and you stick it in 1 to 1 ration usually 4th gear it is close, but an automatic especially with a loose converter, wasting your time.

devil's in the details. what are the engine, tranny, axle ratio specs.

Is the 362 RWHP reasonable for the engine build?

The operator mentioned smart tach. When he explained how things worked I was skeptical.
The engine speed on the monitor was close to the tach in the car but I don’t know how close it has to be.

3.91 axle, Lupo 10” converter behind a 408” small block, ported Eddy heads.

I have seen estimates from some people that this engine packaged should make 450. So it sounds low on this dyno.

Been there, done that. Totally agree.

My old setup made around 425 HP on chassis Dyno and ran 10.90s at around 3450 lbs. that means engine made around 530 HP or so per the HP calculator If I recall correctly.

I’ve seen it many times.

I have dyno tested engines that got put into the car, then tested again on the chassis dyno.
You’d have no clue by looking at the curves it was the same engine.

with 362 at the tire with a low gear and looser converter, which can skewer numbers, the engine is easily making 450 at the crank. an engine dyno can't tell you what's going on behind the crank; only what's going on at the crank. two different worlds. i used a chassis dyno as a tool to help me figure out why my big ideas didn't work. to me it wasn't a numbers game but needed information. now you have the info go back to everything behind the crank and see if those parts work good together. for me the track and chassis dyno answered my questions.

The only way to kind of know what is going on with a car like that is to have an RPM sensor on the driveshaft as well as one on the engine. Then you'll know true engine speed as well as converter slip. Measuring RPM off of the rollers makes the curve useless since there is all kinds of slip between the engine and the rollers.

The test isn't a waste of time, it just doesn't provide a correct power curve. The torque numbers might be correct depending on how they were measured. The test can still be useful for tuning, just ignore the numbers.

With a dyno run like this the torque number isn't correct, it will measure peak torque at wherever the converter flashes, have max converter slip at initial hit/flash and converter slip declines over the duration of the run.

By not correct I mean, it is calculating torque correctly but as multiplied by the converter. Locked converter, manual trans or engine dyno would show true engine torque curve. Or, with an efi car you could see close to where it is by the shape of the VE table.

So, after the start to end of the run, the converter slip is decreasing and 'catching up' to engine rpm, which is why the graph looks the way it looks.

Very easy to see this with a car with a lock up converter.

As an example you could make a run with the converter locked from say 2500 rpm and get a 'normal' looking graph. Show peak torque at say, 4500rpm and peak hp at 6500rpm.

Make the next run converter unlocked and it will look like the graph above. Peak torque wherever the converter flashes and a relatively flat power curve.

Depends on the dyno. Dynojet measures horsepower and calculates torque if it has an rpm input. Otherwise you just get horsepower vs speed.

Thanks for all the replies.

Like I said, after finding out how this dyno looks for engine speed, and does nothing else, I was skeptical that it would correlate to anything very useful except a baseline tune. Performing anything on the dyno on this particular day may have been helpful in improving things but all he was tooled up for was a quick few pulls.
He was just interested in how much this thing made.
It’s his street car, and I doubt he would pull it for a dyno test.

I suggested backing off timing and in doing so 3 times it made it’s max power.

Then left the hood open with a pedestal fan running and that helped to eliminate sucking in hotter underhood air.
Made it’s most power this way.

Let’s say you built a run of 5 engines all exactly the same, or bought them...... like Crate engines.
If you dyno tested them all back to back on the same dyno, the power curves and output should be very close.

Now take those 5 engines and stick them in 5 different cars with different transmissions, gears, converters, exhaust systems etc, and test the 5 different vehicles on the same chassis dyno.

I’m confident the chassis dyno results would be way different from each other than the 5 engine dyno tests.

If you’re really trying to determine what the power curve and output of the “engine” is........ put it on an engine dyno.

Are you sure you don't have that backwards? Because the rollers on a Dynojet absorbs torque. Torque is a measured output (from a strain gauge) and you need a rpm value to calculate horsepower.


Joe

Yeah I'm sure.

What you described is not how a dynojet works, if we are talking inertia and not eddy current.

x2. Inertia dynos measure HP and then back calculate torque with an rpm signal.

Eddy current dynos measure torque.

I learned a longtime ago to use the dyno, regardless of brand or type, to measure the differences in changes, not rely on the numbers given that day as the truth for making performance increases
If you make power and the car spins the tires worst and the ET and MPH go down did you use that information to make a faster car
Remember the old law on actions and reactions

I stand corrected, you both are correct, the dynojet measures work (hp) then can calculate torque with a rpm signal.

Joe