Engine dynos with programmable sweep rate

[Chad Harrington]

Which makes of engine dynos have the ability to finely enough control load to result in an engine speed sweep that matches what the engine actually does on a specific track?  This way the load applied to the engine is very accurate since the engine speed is increasing at exactly the same rate on the dyno as it is on the track.  Programmable sweep rate so we can test for short tracks, long tracks and anything in between.

I’m not referring to engine speed “sweeps” that many water brake dynos do which is actually more a step test than it is a real sweep (eg. increase engine speed in 1000 RPM increments, holding at each step for half a second to record torque at that rpm before reducing load and climbing to next RPM point, holding it there, and repeating).

Also not referring to lap simulation via throttle servo control.  That’s a nice feature, but secondary to being able to specify a sweep rate that matches what happens on the straights at a specific track.

Also looking for automatic closed loop water temperature control so we can simulate cold or hot conditions.

Looking to buy either used or new.

Thanks a lot guys.

[John Matthews]

Really?

For kart motors?

Do tell why….

[Chad Harrington]

John Matthews wrote:

Really? For kart motors? Do tell why….

Because you want to replicate load and know for that load range exactly what ignition timing and carb settings make the most power and power width.

From data we know that kart acceleration is not proportional to gearing.  In quite a wide range, you can go taller on gear and not lose any significant low end accel, yet gain massive speed advantage up high.  So we know that not just engine speed but engine load, and/or carb sensitivity to engine acceleration (engine speed derivative), are key factors in kart engine power width (at least for KF2) .

Walt, thanks.. when you change gears on an inertia dyno and overlay on-dyno engine speed vs on-track engine speed, do they line up perfectly?  Do you mind showing some examples?  Really appreciate it.

I tend to think the on-dyno engine speed trace will be closer to linear vs the on-track one mainly because of air drag increasing as a sqaure of speed, but perhaps in karts and at typical kart track ground speeds, it is small enough not to matter.

[Walt Gifford]

On an inertia dyno all you have to do is change gears being mindful of the wheels max speed of course.

Gif

FAA certified jet engine and aircraft technician,
Nicholson Speedway class champion 2001,
Yamaha KT100 Service Center,
41 years karting experience

[John Matthews]

I guess my question is do the engine manufacturers even have such a system? I know they exist at the F1 level but they have much larger budgets than most kart teams.

My dyno is old and home built without any automation or data logging at all so I’m definitely out of my league when it comes to programmable/automated load simulator deals. I just have a big knob you crank down until you get to the max load whatever engine you’re testing will handle. With experience I can set timing and carbs pretty close to optimum but without having the actual air that my customers will be using on any given day at the track it’s never gonna be perfect. I personally view the dyno as my tool to make sure the work I do is having the desired result (I mostly build 4-cycle open motors), and that it’s not going to break first time out.

I understand racing is about getting the absolute most from every element but it seems like there is a point of diminishing returns when it comes to tuning on any dyno. Perhaps a less sophisticated system would get you 98% of what you are looking for and you could spend the other $50k on testing at the track?

JMHO,

John

[Rodney Ebersole]

Chad, With an inertia chassis dyno the “sweep time” is controlled by the final gear ratio on the kart and the power out put of the motor and chassis. Simply put if you make more power  or change the gear ratio the sweep time will change,  just like on a segment of a straight track.

If you wanted to test  longer or shorter sweep time the karts gear ratio would have to be changed until you mimicked your desired sweep time that you have predetermined to be the sweep time for your testing. That also eliminates the need to mimick arrow drag because your on track sweep time all ready includes the drag so the karts gear ratio would have to be changed from on track gears to the proper dyno test gear that makes the sweep time the same for booth types of recording.

The only other method that could be used to change the sweep time on an inertia dyno to match on track sweep time would be to change the inertia load or weight.

[Chad Harrington]

Thank you guys for your sensible advice.  I agree on diminishing returns as we dive deeper into detail, but also want to look at how to maximize the money spent on an engine dyno.  I guess what I what I was hoping was a fast control engine dyno that can add the exponential aero load, without costing too much.  Since it is not rapid transient like track sim or for very powerful engines shooting through low gearing, but rather the opposite and just optimizing for the medium and long straights; such load control should not be difficult to create.  After all we can gear to add most of the aero load after best linear fit, and be very close in a specific range.

James, the data to simulate the load is easy – the straightway acceleration engine speed curves on track and dyno, just have to match closely.  It can be a 2  – 3 try iterative process, starting with calculation of the engine load, which will get close and make a good starting point.   Thanks for the aero numbers… how long ago were they reported?  Is it for current generation bodywork?  I plugged the numbers you provided and got the drag force shown in red.  The dashed line is target additional inertia load I would gear for (artificially tall) on an inertia dyno for best fit in the speed range that matters (most common medium to long straight speed range).

[Chris Cooper]

Hi Chad,

If you’re looking to have a dyno that simulates track loading, you’ll need an adjustable load source.   All of the above are great reference for the use of an inertia dyno (or water brake) at a fixed ratio (i.e. fixed load).  To simulate what an engine will see on the track will require the use of a fine adjustment.  This can be done in a variety of methods:

– closed loop variable frequency drive on a std AC induction motor (will need a brake resistor to handle the regenerative energy)

– a true AC servo drive (again – will need a brake resistor to handle the regenerative load)

– a proportional relief valve if using water or hydraulics (insure you have enough fluid capacity to handle the effects of heat rise)

A small programmable controller that monitors axle speed of the dyno and engine speed should be able to give a command signal to the drive(s) or valve in proportion to engine speed.  It will take a little while to map the command signal to the engine speed, but it’s all very easily done.  Once you figure out what the command signals(offsets) need to be, you can adjust when the fine adjustments are made during a pull.

Hope this helps,

Chris

 

 

 

[Chad Harrington]

Thanks Chris, I’m very bad with electronics so will never be able to efficiently create a control system like that.  I will need to buy an existing solution to save time.

I’m not actually trying to run track simulations per se.  Only trying to simulate the 2 or 3 most common length straights, including aero load especially the higher speed straights, since aero load is the only load that increases exponentially with speed.  The linear loads can be adjusted with gearing on the inertia dyno.

[James McMahon]

I think my question might be how are you going to capture the data that you might use for the load curve?

For aero…. I read a study a while ago that used a combo of CFD and other testing to quantify the effects of it. In short, a typical CIK bodied kart has a CoD of .8 and a frontal area of .570 sq/M (6.1 sq ft)
So you could use that as a base to see how much, if any aero you want to factor in for your testing/simulation. I guess it would depend on the average speed around the track, more-or-less.

[Author]

Posts