Kevin

Geoffrey, what is the MAX speed /MPH that the Dyno Dynamics can run??

TB993tt

Geoffrey, what is your take on the loopy peak torque figures these dynos show - my take on the DD is that your Chevy will not slip 'cos the rollers don't load up like the Maha but they can produce silly numbers like the 870NM (real ~720NM) 996 GT2 above

Geoffrey

The DD will hold the engine at peak torque for an infinate amount of time, so it loads the engine as much as you possibly can, enough to overcome the torque it produces, whatever number that ends up to be.

To me, I don't really care, as long as the dyno can hold a load and I can tune an engine for peak power, the dyno serves its purpose regardless of what numbers it produces.

TB993tt

It is strange that the dyno can be loaded like you say but on the much vaunted (by DD) "shootout mode" which they sell as the gold standard for measuring hp the rollers do not load like the Maha and give the spurious readings I reffered to.....

Maybe they decided they concentrate solely on the peak power reading and do without the problems caused by loading for peak torque by compromising, letting the rollers calculate the real torque by some formula rather than an equivalent load ?

Geoffrey

I'm not really sure what you are describing because I don't see that as having anything to do with the capability to hold a load. I've used dynos for the past 15 years and the DD has always worked the same way. You can tell the dyno to hold the engine at a certain RPM (or road speed) and regardless of how much torque the engine makes within 1ft/lb. At peak torque, that is as much load on the engine as you'll ever see, so if the dyno keeps control of the engine, then it is properly loading the engine. You can also program the dyno to run the engine through the swept run test over a certain period of time, so that loads the engine based on the dyno's control over it. I believe the DD does so similarly to the Maha and I know it is similar to the Dynapack. If they produce different numbers so be it.

I'm not subscribing to your theory that dyno manufacturers have ulterier motives. Both the DD and Dynapack dynos are great dynos to map engines on and can very accurately control the engine function. I just find it funny that you are making ascertations about a product, or a geographical group of products you have never had experience with.

Woodster

This is a very interesting read and I am amazed at TB's willingness to subject to his
car to so many tests.
I recently witnessed my first Dyno. A modded 996GT2 on a DynoJet. I thought it was
nothing short of comedy (especially after hearing some of the other numbers
that were produced on it and the motors that accomplished such numbers in a 10 second run.
After all the dynospeak, I do not have ANY idea if my 628 RWHP make sense or not (on a DynaPack).
I know Todd Knighton's reason for purchasing it was not dyno queen bragging rights but
to be able to load and tune a car (He used to do this on the roads!!!! At dangerously high speeds).
thanks for all the thoughtful insight and discussing.

Marty K.

TB993tt

What I am describing is simply the DD's ability to measure big torque using a lightly loaded roller (the 870NM WM example) compared to the Maha loading up so much during a swept run that measuring over 750NM becomes challenging and causes tyre slippage -

I am guessing that the DD (maybe intentionally) does not load the roller as much as the Maha during the swept run at peak torque and extrapolates torque using proprietry formulas compared to the Maha which directly applies a matching torque (and hence the tyre slip issues) -it is the DD's extrapolation which does not seem to produce accurate peak torque numbers (for a swept run)
My "experience" is by looking at a DD swept run "shootout" chart showing 870NM for a ~720NM car and many other similar over reads.

If you read the DD web site it is clear that high on their "motives" is to make their users money (a fair objective) and it does not assist their customers
when they can't measure high torque turbo cars because of slippage problems hence they may well have built in what I describe above to ease the measurement process (I am trying to imagine one of those 1000hp ricers on a Maha - would not work)

TB993tt

Here is one I dug out. I think this is the Sportec Bosch dyno which mw911 refers to above.
I think it is a 993tt engine - the peak torque number looks far too high whilst the power is definately achievable.......


Here is another Maha (can't remember where its from) of a 996GT2 -it manages to translate 430.7PS at the wheels to 598PS at the flywheel - check out the "Drag power" losses of 181PS !!! this is double what my 993tt measured -this is the number which can be manipulated by an unscrupulous tuner (who wants to market his GT600 kit -or whatever ) touching the brake during the coast down losses measurement.....
Again on this one the torque is measured within the scope of wheel slippage for the Maha ~748.5NM
This is actually a nice curve IMO and will be a fast car with good torque spread and flywheel 530PS/750NM genuine

francisb

for what its worth, the more i learn and read about chassis dynos the less use i think they are except for bragging rights. it seems the operator can produce whatever result they want, there are just too many variables. my very modified impreza, s4tt and (standardish) 993 have run harder, faster and smoother when mapped 'on road'. there's no substitute for your mapper and tech risking their lives to map your car properly!!! :-)

eclou

TB,

I think it is very hard for me to accept the position of the MAHA being the only "accurate" reading in the face of the revelation that it cannot even hold traction at higher levels of torque. In fact, being that all the other dynos seem to have little problem holding traction at similar power levels (intertia and loading dynos) it would logically follow that the MAHA is not necessarily more accurate but rather has had a limitation exposed which would lead one to throw it out as a valid data source when testing over 700+NM. The evidence you have shown us now suggests that the limitation is more of an operator/setup error than an inherent design flaw - they simply are not strapping the car down sufficiently. Would anyone here allow their car to be transported on a flatbed trailer with only a single cross strap?

I do not believe that we can conclude that the slippage is because of the "unusual loading." For there to be traction loss that loading would have to exceed .9g for street tires which is not something you will ever see in 5th gear unless you generating the kind of torque the 1000+HP Supras are churning. That kind of loading should also be enough cause major driveline damage such a twisted axles, blown CV joints, etc. It would be similar to subjecting the driveline to a load encountered when launching a car from a dead stop, but holding that load constant. That seems to be very unlikely.

Now let's accept that the loading is real and it can cause traction loss at 750 NM - of what value to judging real world power? Under what conditions could you make your car spin the wheels accelerating in 5th gear on the highway? Perhaps full throttle while under full braking? I would hope that my car never was subjected to such a scenario, and that tuning a car to such a scenario would not be optimal for the real world conditions it would be driven in

Geoffrey

Toby, the DD uses an eddy current brake just like the Maha and, up to its torque rating can be used to load an engine sufficiently to overcome the engine torque and hold the engine's load. I've done it, and done it with big block American engines that produce more hp and torque than a Porsche turbo engine. The base DD unit has enough brake capability to hold any 993/996tt engine we're talking about here. If you select a run that is quick, say 3sec the dyno will not have to use much of the brake (lightly loaded), if you select a long run, say 30 seconds, it will be required to use a lot of brake to hold the engine from running away. Now, this is where I think the DD PID algorithm is excellent, I have found, compared to other dynos I've used (dynapack not included) that it can very accurately control the engine load or torque. As a comparison, if you do the same thing on a Mustang dyno, the Mustang's PID algorithm is not accurate enough to hold an engine within less than about 15-20ft/lb of torque. It is not adequate, and I spent 3 days at Mustang with them and their Viper test car trying to get it to work properly. It does not.

As for outright numbers, I don't know and don't care. I'm only taking issues with you saying that a DD is lightly loaded, or that it cannot load the car properly, that is an incorrect statement from someone who has never ever used that dyno. That is not the case with the dyno. You can take issue with the numbers all you want, I don't really care, but to say it is cannot load the engine is an ignorant statement.

Maybe Maha will decide that it really wants to sell dynos here in the US and will actually bring a real dyno to the PRI show this year. Last year, when everyone else had a dyno outside the convention center with a car on it, they brought an emission dyno capabile of 50hp. That was a joke.

eclou

Geoffrey are you using this:

http://www.dyno.com.au/dyno/controll...owAWD450DSPage

I see there is a quad retarder version as well. Interestingly enough, I found there is one in Houston! The specs read that it is using coupled rollers so there should be no undue stress on the AWD differential correct?

TB993tt

Geoffrey
You seem not to be grasping what I am trying to say......

I am not saying anything about the DD's ability to do all the things which you claim above - I have never used the words "cannot load the car properly" - you have said it can and of course I believe you

I am saying that for the DD "shootout mode" swept power run the DD does not load the rollers as much as the Maha when the Maha is set for "ott-Motor/Turbolader (luftekuhlt)" the Maha has a specific setting for testing aircooled turbo motors (Porsches !) does the DD ? From what you say above I suspect not...
Maha have obviously spent time coming up with this "program" for our cars but it has its limitations at mid range torque in excess of ~750NM
The DD spews out incorrect peak torque numbers for (tuned) turbo Porsches when in "shootout" mode - its peak power figures appear reasonable...

Eclou
I am not saying the Maha is the only dyno capable of accurate power reading, it is teh mid range torque readings of other dynos which I am disputing (and more accurately the mid range torque of modified engines)

I really can't get into the argument of how much G etc on a dyno vs street 'cos I don't understand the forces involved - I do know that having spoken to a Porsche racer in the UK who has had many high torque cars on his Maha that they don't attempt to measure big peak torque engines 'cos the tyres slip no matter how they tie it down -they use part throttle and then WOT after peak torque......

Geoffrey

Eclou, I've used the single and double retarder 2WD and the double AWD versions of the Dyno Dynamics dyno. I've not been to a facility that has the quad retarder and I've not exceeded the load capability of the dynos. I have exceeded the capability of a Dynapack 3000 (smallest version they make) with a turbo 3.8l Porsche.

Toby, I do grasp what you are saying and will suggest that if the run on the Maha set to Porsche testing mode takes 20 seconds for your car, and you program the DD for a 20 second run, they will be loaded the same, ie the dyno is controlling the engine's ability to accelerate by loading it. I don't use the shootout mode of the dyno, and it really isn't mentioned on their website except for a small box in the specification section of the dyno. The DD does not have a Porsche turbo specific mode, but the DD does have the ability to create user modes, although I've never used that fuction. I believe the shootout mode exists for specific comparision used mainly in AU.

Here is a question for you to ponder, If HP is torque x RPM / 5252, how can you have a result where the HP is "reasonable", but the torque is not? No dyno measure HP directly, it is calcuated from the torque measured at the load cell (whatever design they use).

TB993tt

The torque number being "extrapolated" by a lightly loaded dyno (one which does not load enough to match the torque) during a "shootout run" is calculated by proprietry formulas (based on, and accurate for, tests with stock cars I guess) which do not mearure the torque correctly at the mid range. For some reason they seem to "settle down" once the roller has been violently "spun up" and the torque measurements higher up become accurate

It is something to do with the way a high torque tt engine accelerates unsufficiently loaded rollers, the formulas used (by the likes of DD, Bosch) can give accurate torque readings so they do indeed work for most stock cars, but not for modified high torque tts.....

The Maha seems to attempt to brake real torque and fails over 750NM - for sub 750NM the Maha gives great and accurate power curves (providing the temp and drag losses aren't fudged )