Perhaps the countless runs done on AWD Dyno Jets with out a single problem reported could be considered proof that it is not an issue?
I feel that doing 20 let alone a 100 dyno pulls in a row (with only "20-30sec cool downs") on any chassis dyno is IMO foolish and irresponsible. You don't really run cars on your dyno like that do you?
I don't drive on chassis dyno's, for that reason most of my tuning is done on the street/track. Even on the street I perform much longer cool downs between pulls and with considerably more air flow than you get on any chassis dyno. Doing 3-4 reference pulls (with at least 10 min cool down between runs) every once and awhile on an AWD Dyno Jet is certainly not going to damage an EVO. I think even you realize that, otherwise if they are so dangerous why do you still use them on your out of town tuning sessions?


Really? And you can prove this because? Obviously you've never seen me on a road course. The difference in load sliding around a corner is so great that the center diff is unable to maintain matching speeds fore/aft like it does on the dyno. So how can that be less than on the dyno? And maybe it is for a slightly shorter period, but it is for corner after corner after corner. On the dyno it's one 10-12 second period and then hopefully a good cool down.
Can you see why your arguments just don't add up for me?

Other than you’re off the cuff claims, I have not seen any supporting test data. Let alone verifiable data from a credible independent source, what am I supposed to think? Just because you throw out a few big company names I am supposed just change my entire point of view? Get real.
Your theoretical arguments have been weak, and your supposed factual claims have no verifiable supporting evidence. But I am the one being unreasonable?

So does this mean your not going to answer my earlier post questions? You obviously misunderstand me. If I was absolutely certain that I was right we would not be having this conversation. With such strong opinions on the subject I figured you might have some stronger theoretical arguments and/or supporting test data.

Here are some additional thoughts to consider:
The logic behind these statements is hard to argue with.
(most of this is from the Dyno Jet web site)

inertia dynos are by the laws of physics; are the most repeatable engine performance testing device available anywhere in the world. The reason for this is that an inertia dyno’s function is purely mechanical. In other words, since the engine’s resistance consists of solid steel discs that will not change during the test run, the possibility of variables are practically non-existent. There is no chance of any variable due to improper temperature control of hydraulic oil affecting the viscosity, “dirty” water or electricity, improper valving of intake and/or exhaust water, etc. The simplicity of design is the strong point here.

An inertia dyno is the most advantageous tool for race engine development when the engine builder seeks to monitor and improve the speed that a racing engine accelerates. This is due to the fact that the engine is accelerating against (and therefore controlling) a constant mass, which is comprised of the inertia wheels. On all other dynamometers, the “absorbing” unit controls the speed of the engine.

The inertial dyno can only measure and readout hp during acceleration, while a braking type dyno can measure hp continuously.

Well then, let's always measure continuously! Not so fast. The brake on a braking type dyno has to absorb all that horsepower, which means it gets HOT! A typical 100 Watt light bulb gets pretty hot. Image a brake that is absorbing 400 hp. That's 299,200 watts, or nearly 3000 of those one hundred watt bulbs in one spot. The brake dyno has to get rid of that heat, and generally can't do that for a long time. For continuous hp measurements, typically the engine is out of the car and mounted to a dyno that can dissipate the heat continuously. So that gets us back to fairly short runs with a brake dyno. You can however run for some short time at a fixed rpm (or horsepower) to analyze some facet of the engine's operation

In both types of dynos, rpm is very accurately measured, to fractions of a rpm. No problems here.

In the inertia dyno, the inertia of the drum must be measured or calculated. This can be done to great accuracy, and the reference article cites the inertia is measured to 4 digits. For a pair of 4 foot drums that weigh 2700 lbs each, that is about one part in 5000. Very accurate.

In the brake dyno, the measured item is torque. You either have to measure the actual torque in the shaft of the brake with a strain gauge torque cell, or measure the force on the brake mountings with a strain gauge load cell. This can be done very accurately, but not to the precision level of the inertia calculation. I found load cell accuracies on brake dynos to be in the range of ˝ to 1% stated in their websites, but no overall dyno accuracies were stated.(Remember that's 1% of full scale, so if it's a 1000 hp dyno, it can't tell the difference of 10 hp)

If the brake type dyno has a load or torque cell to feed back the torque settings, it can only be as accurate as the accuracy of all of its parts. And I would add to that the accuracy of the operator programming the system.

Kind regards,

Eric

 

Dyno Dynamics dynometers are calibrated at the factory, what I mean is they build the parts, assemble them and calibrate the machine, and then disassemble them and ship them out. But there are some variables the owner of the dynometers control, plus other variables that no one can control, such as weather. In other words, no 2 dynometers in different locations will never read identical.

 

Before we get into this any further, could you please answer the following question regarding Dynojet resolution and accuracy:

What is the cause of those jagged, seemingly random, high frequency dips and valleys that we see in just about every dynojet graph (especially when the "smoothing" feature is set to the minimum)? Does wheel output actually jitting up and down like that during a run? Or are we seeing, among other things, a measuring/sampling artifact inherent in the way the Dynojet measures output? And why should there be a smoothening feature to begin with? We're talking a high-precision measuring device here, right? Kinda defeats the purpose of measuring something if your connecting the dots with a "best fit" curve afterwards.

shiv

 

I think I have answered plenty of your questions, while most of mine still go unanswered. Oh and DD won't send me squat, perhaps you would like to share?

But just so you don't start accusing me of being evasive again. Yes it is measuring/sampling artifacts inherent in the way the Dynojet measures output. I believe most of that is due to slight inconsistencies in the engine RPM pickup. It is a minor variation that does not introduce large errors in the data. Systems like yours do not rely on the actual engine RPM for the calculation, that could explain why the power/torque peaks come out wrong occasionally.
Just because you cannot adjust it does not mean there is not an internal software filter on your system also. I would be very suprised if there were truly no filters, not to mention that I have seen some of your plots looking a little jagged in spots also.

Once again that proves nothing, so how about my questions or maybe send me your manaual? Does it really have all the answers I seek?

Kind regards,

Eric

 

Nope. Those artifacts exists without the tach pick-up as well (that is, just going off of wheel speed).

When you can't distinguish between misfire or pick up the exact RPM point of knock sensor activity, this "minor variation" (and when is a 5ft-lb dip minor?) causes problems. There is no excuse for sampling noise in a measuring device. Especially when it interferes with tuning or diagnosis. But it can lead to higher numbers since it tends to elevate the peaks a bit.

No. The Dyno Dynamics Dyno does rely on an inductive pick-up for computation. Of course, it offers 3 levels of pick-up sensitivy which makes means that you can get it to work with every ignition system (coil on plug, plug wires, etc,.) and not suffer from the same break-up problems you see with the Dynojet pick-up. For cars that have their ignition system buried under engine covers (making them hard to get to), you can get an estimated engine speed by going off of wheel speed and a gearing factor. When doing this, you do lose a little bit of accuracy since it does not account for tire deflection (and the resultant rolling diameter changes) that occurs during the power run. If the power/torque peaks come out wrong, it is because the user didn't calculate the gearing factor correctly. It's quite easy to do, in case you're wondering. Simply limit wheel speed to, say 50mph. Go WOT and look at the XEDE or a scan tool to see see your engine revs. In the case of the EVO with factory wheels/tires, it will go to 4600rpm, giving a gearing factor of 92RPM/MPH (4600/50). It's hard to mess it up. But it's possible, I guess.

I did a demonstration last year at a dyno day showing the resolution of the dyno for those who were used to seeing jagged and noisey Dynojet results. I mapped a TEC3'd turbo Miata to have have a big ignition advance dip during a very small (10RPM) span of engine speed. Basically, I mapped in a 4000, 4010, 4020 and 4030 RPM break points. Between 4010 and 4020RPM, the ignition advance dipped 5 degrees. On the dyno, there was, as one would expect, a big 15ft-lb dip in torque during that narrow interval. I did this test on a Dynojet only to find that dip was nearly indistinguishable from the general noise. And smoothening out the curve erased that dip entirely. So contrary to what you said earlier, this noise and sampling errors do induce a significant error in data.

FWIW, this is why it takes me so much longer to tune a car with a Dynojet since it's hard to pin-point the *exact* point of knock correction activity.

And I do my best to tune that out

Cheers,
shiv

 

Here is a link to KTR Performance's FAQ page pertaining to their Dyno Dynamics AWD machine. Don't know if it will shed any new light on any of SilverSurfer's questions or just fuel the existing controversy, but it is another source for information. FAQs may not reveal anything new, but KTR may be willing to answer specific questions.

KTR Performance Engine Power and Dynomometer FAQs

KTR Performance Dyno Dynamics Results for select vehicles:

Stock 2002 Z06 Corvette: 323.7 WHP
Stock 1994 964 Flatnose 3.6 Turbo Porsche: 305.2 HP

 

FYI, I was at Axis Power Tuning to day and got some baseline runs on the new Dyno Dynamics dyno. It's completely stock, temp was 85 degrees, humidity was high, rained all day here, it was at 85%. My average was 204 whp, the most I got was 212 whp, the lowest was 198 whp. The shop is 50+ miles away from my house, I didn't drive the car hard on the way there, but I was on the dyno less than 30 mins after I got there. My dyno sheet shows how protective the factory ECU is, running rich at high RPMs and with only 30 mins of cool down time, it was still pulling timing. Peak HP was at 6000 RPMs, not 6800. After 6000 the graph got rather irratic after 5600, showing +/- 8 HP every 200 RPMs. I think with a longer cool down before would have gotten me a few more HP, probably around 220 or so.

 

198-212 wheel hp sounds right for a stock EVO running 93 octane. On 91 octane, we typically see output in the 195-205 wheel hp range with a minute cooldown between runs. Increasing cooldown times to 2 or 3 minutes doesn't have much, if any affect, on output. Of course, a lot has to do with how much airflow the dyno's cooling fans put out (for proper radiator and IC cooling). The weaker than fans are, the more cooldown time is needed. I take it Steve Nichols was out there today for the training session? Great guy.

Cheers,
shiv

 

Oh,, where to begin, ok first take a look at the actual roller speed measurements that I posted on this thread, they look smooth and accurate dont they?. While your dyno uses the ignition pickup in it's calculation, it is not an intergral part of the calculation. That is why you can plot power/tq without even looking at actual RPM but estimated. On he Dyno Jet, you lose the ignition signal, you do not get a reading.
BTW here is one of my typical EVO dyno plots done on a AWD Dyno Jet, with no smoothing and scaled to look similiar to your plots. From what I can see it does not look all that different from yours. This is with a stock ECU and an MBC that fluctuats boost a little more than the stock BC.
Most Dyno Jet plots are scaled in such a way that these minor output fluctuations are highly exagerated when compared to your graphs. I have also found that even slight fluctuations in boost and/or ignition/fuel irregularities are easily seen on a Dyno Jet so..
If you get a good ignition source they are very accurate.

 

her is the same graph with max smoothing, I do not think there is any significant loss in usable data, do you?
Oh and the fluctuations that I believe may be erroneous are 1-2hp fast little dips/peaks. I feel the same way about similiar dip/peaks that I see on your plots also, although there much harder to see on your graphs due to the scaling.

 

Oh my, let's just take a momment and look at the variety of test data from this super accurate dyno. I don't know about you, but I am having a really hard time trying to convince myself that any of this data is accurate. They all seem way off, but the best one for me is the two (A4/GTI) with very similiar powertrains. One FWD the other AWD, and the FWD has greater drivetrain loss?

 

Sorry Silver Surfer, but you are losing info with the smoothening feature on. And that feature exists, in the first place, due to artifacts/measuring error caused by the Dynojets sampling rate.

The real question is which graph is more accurate of real wheel output: smoothed or unsmoothed. That is, the unsmoothed does indeed have less info loss but it also has more noise. But the smoothed has filtered away the noise and some legitimate info as well.

If you look at other Dynojet results that have been posted on this forum, this becomes even more evident. The fact that many of the results, like yours, have areas where the curves dissappear and come back (due to tach pick-up loss) suggests that there may be a better way to keep track of engine speed as well.

And, in that Miata test I described earlier, why didn't the Dynojet pick up that transient torque loss? It was there. It could easily be felt on the street and seen on another dyno (it felt like a short series of misfires) but it never showed up clearly on the Dynojet. You have suggested that the dynojet's simplicity yields superior accuracy and resolution. However, the results of that test would suggest otherwise. From my own perspective, I've used Dynojets for years. They served their purpose well when doing power runs. They worked well on low to moderately powered cars and no-so-wheel on high powered cars (the tires would slip off the roller during spool up). But as a tuning/analysis tool, they didn't offer the resolution for the user to be able to distinguish between normal dynojet noise, actual misfire or knock retard activity. Let alone all the normal shortcomings of inertial dynos (no ability to load steady-state, adjust ramp-up depending on engine ouput, control fore-aft wheelspeed (in AWD applications) and so on.

It also seems that most of your experience with brake dynos is from the Mustang Dyno you often refer to. While this is indeed a loading dyno, it offers a very limited sampling rate. Last I checked, it only samples 20 or 30 times during a dyno pull and connects the dots afterwards with straight lines. This is why things like torque spikes during boost overshoot, misfire, knock activity, etc,. never show up on the graphs. This is also why peak hp and torque is only registered in 250RPM increments.

Cheers,
shiv

 

Same day, same car, and a 14WHP difference! You actually believe this is accurate?
I have done back to back runs on both the Mustang and the Dyno Jet (three run's) with only a 2-3 minute cool down, at the most the HP varried 2-3 WHP, (Mustang was >2).
Shiv is right in that longer cool downs do not seem to make a significant difference in power measurements. I usually allow longer cool downs to protect the drive train.
No wonder you are seeing such a large deviation in HP from the same make/model but different cars. With minor (less than 5% IMO) tolerance variations from car to car, enviromental changes, and this dyno, no wonder!

This is a nice dyno for for various reasons, (packaging, flexability, etc), but as more and more data becomes available, the overall output accuracy just does not seem to be there.

I really suspect that after absorbing so much power the electronic load cell's start fluctuating. It's just a theory but it seems reasonable to me based on my knowledge and experience with similiar circuits.

And one more time in case you missed this:

 

Here's an email I got from Dyno Dynamics this morning:

---------------------------
Dear Shiv,

Steve passed on the link to the EVO forum which I read with interest.

First of all ShootOut mode is not just for flywheel power calculation. Although it does a super job of that. Much better than Coast Down.

The prime objective of ShootOut is to create uniformity of readings from all DD dynos. This is done by causing uniformity of dyno settings (this is done by the software selecting the most appropriate drivetrain inertia and ramp rate for the class of vehicle being tested). The rest is acheived by training our customers (eg. you) in the correct methods of tie down, tire pressure, fan positioning, good ventilation neccesities etc, the list goes on. If you set up the car and the dyno the same and the car is repeatable, the readings will be repeatable, world wide!

One of the members doubted the reality of our readings, because our atmospherics are entered manually. That is why ShootOut prints these manual inputs on the dyno graph. If the operator enters figures that are "slightly" inaccurate, the power variations will be extremely minimal. If howeverthe operator inputs unrealistic figures large enough to significantly effect the power readings, then this will be obvious to anyone who studies the graph.

The principal of ShootOut is to ensure that there is no difference between your dyno, Dyno-Comps dyno, Axis's dyno, or any other Dyno Dynamics dyno run correctly in ShootOut mode.

There will always be some small variations in readings because of the difference in locations therfore conditions that can effect the way the engine management system behaves and it turn the turbocharger system behaves (their behaviour is inherintly linked). No dyno can automatically compensate for this.
----------------------

 

Ah, see now I do not disagree with anything you stated above. For tunning and diagnosing I would much rather have your dyno. My whole point is that for measuring absolute max acceleration WHP and then comparing those values with other dynos, inertia dynos are far more accurate in that regard. It sounds like you might agree to this point, if so, I think were done?

BTW coast down measurments show about 40-45 HP loss at 6-7K on the Dyno Jet (don't have the graph yet). The average EVO seems to show around 230 WHP on the Dyno Jet, these numbers add up nicely with the factory crank HP.

Kind regards,

Eric

PS all of this chassis dyno talk has gotten me to researching and analyzing drivetrain losses (which I think we essentially agree on). I plan on starting another thread to discuss this in more detail.