Self Built RC DYNO

[Fr0sty]

Ok i have been reading and following (its about all the time i have lately) and seen the issues there with Win 7 so i thought i would test it as i have a few around the office.

On my laptop running Win7 it works no problems as long as i run it in administrator mode, however on 2 out of 3 HP desktops running Win7 it pops up with the "Initialize wave input error".

The 2 machines that error are both running Kaspersky Anti Virus/Hacker/Spam etc etc dont know if this is related as i can not uninstall them.

If you want me to test anything for you just let me know

PS i really think this thread should be moved to tutorials to increase its expose, with everyone's input this is now a viable project

Really love the Dyno in its state now dash as well awesome (hack) with the multimeter, and carnt wait for ver 3 of the SimpleDyno software you really are putting some effort into this and it is appreciated here

[DamoRC]

Fr0sty - thanks. This is really helpful. It appears that the problem is not OS specific but is probably audio hardware related and how audio is handled since the release of vista.

Any chance you could tell me what the audio setups are on your lap top and the other ones you tested (sound card info might suffice). I think that if I add code to allow you to select what audio devices to use, it might help.

DamoRC

[Fr0sty]

Fr0sty - thanks. This is really helpful. It appears that the problem is not OS specific but is probably audio hardware related and how audio is handled since the release of vista.

Any chance you could tell me what the audio setups are on your lap top and the other ones you tested (sound card info might suffice). I think that if I add code to allow you to select what audio devices to use, it might help.

DamoRC

Ahh now that would make sense as the 2 that work (The laptop and a Dell Optiplex) are both onboard sound chips using ADI soundmax drivers (Via chip set i believe)

The 2 that dont are both HP 5700D running realtek soundchips and drivers

Anything else just shout

[DamoRC]

Thanks again Fr0sty. I will continue to follow this avenue of inquiry.

In an attempt to help - I have uploaded Version 3 to the website. Go ahead and try this version out. There are some minor tweaks to the input side of things that might help.

Dash - any luck with the workarounds I suggested?

DamoRC

[-dash-]

@Damo Not had chance to have mess around with it today. wanted to have good few hours to mess with settings and laptop.

soon as got laptop on no wait need to pop out and see a mate then phone and list goes on lol.

going to have proper sit down tomo and see what happens look forward to trying V.3

@Frosty starting to wonder where you got to. multimeter thing was easy but thanks anyway. still want to re-make it tho its the tinker in me.... when get everything else done list is getting smaller tho

[DamoRC]

Dash - looking forward to hearing how it goes. The main thing is to see if it will start and stay running Also, see if the signals turn up in the window "real time". This version launched and ran no problem on my Vista machine (unlike the old version). But there was a big lag in the signals which is no good (and was not solved by the suggestions I made earlier) so I have added a couple of tweaks. The line input slider is disabled in the version I put up because it was crashing Vista - no point in adding more problems to the ones we already have.

[Fr0sty]

Ok take advantage of being a manager for an IT team.

Had them testing the v3 on a few different Win7 machines and it seems it only comes up with the wave input errors if there is no active Mic (Line input).

This is completely repeatable so id say its pretty solid, With MIC ON and active it runs fine and shows the sound level, With MIC OFF or Muted it crashes with the wave input errors.

This is the same for v1.2. So i would say make sure that the sound line in is enabled and working before you launch the app

Few things ive noticed on V3

The display of the sound level is a little delayed seems a slow refresh

If the sound input goes of the scale it crashes the app on windows xp or win 7 (I know its beta) just throwing bug reports at you

@dash

Yes the idea of the multimeter is simple but its the fact you thought of it that's cool. true Hack a Day spirit

Yes not been to active of late, well been to active i suppose is more right.

Haven't had any time to go near it but i subscribe to this thread so i get the email on a morning on my Iphone read them then think i must reply when i get to work but never seem to get chance.

Where are you guys from because you always seem to post early hours of the morning to me i am UK

Edited July 2, 2010 by Fr0sty

[evolution]

Hello DamoRC,

Today I download SimpleDyno program. I have some questions:

- what You mean with " frontal area" ?

- and what is "drag coeff." ?

My rollers are solid ally, so I put only rollermass into program ?

Tia, Evolution

[DamoRC]

@ Fr0sty - excellent feedback. I should be able to modify so that it checks for the Mic status before trying to open it. The "crash when signals go off scale" is an easy fix that I just haven't gotten to yet.

The big problem is the slow refresh. It is consistently slow on my Vista desktop, but 90% of the time it is fine on my XP laptop. I think that this will be trickier to solve and may rely on users configuring their sound setups a certain way (although I don't know what that "way" is yet). I believe that the newer advanced audio setups are getting in the way. Will keep working on that.

By the way, I am in San Diego, thus the odd posting times.

@evolution

Sorry - this version really should not be up yet as I haven't put together instructions and , as you can see, there are still some bugs.

Frontal area is the height of the car in mm mutiplied by the width of the car in mm. You can put in 10000 here to start off. The drag coefficient is the drag coefficient of your car (just plug in 0.5 here). These are used to show you how much power is being lost to air resistance as the car increases in speed - which helps to give you a more accurate indication of what your top speed is. The only numbers you really need to input accurately are the roller mass, diameter and wall thickness. If you are using solid rollers put in half the diameter into the wall thickness box.

Evolution - is the software working for you? What OS (vista, xp, win7) are you running?

[evolution]

Evolution - is the software working for you? What OS (vista, xp, win7) are you running?

At the moment I don´t know. First I must made a coil and a magnet on my rollers. I use XP, Vista and Win7.

Thanks for the info.

Regards.

Edited July 2, 2010 by evolution

[-dash-]

well just had littel play with V.3 on the desktop works fine far as i can see. only crash was when forgot to turn mic on my headphones off and banged the mic (thats proper headphone/mic for games vids etc not hack coil)

few baby spikes +50 rpm when rolling is running down when plotting info but sure get that sorted with tweeking.

only other thing is torque graph cant seam to change the scale allways spikes of the chart.

cant see any of this lag that talking about.

well only when opened V.2 and V.3 at same time to compare this rpm spike thats when V.3 started to lag out.

right now time to plug laptop in see what happens.

and yup im in UK

[DamoRC]

@ evolution. You can run the software without having built the dyno. If your mic is on you will see a wave in the black box and if you talk or tap the mic you will see the wave react, this will tell you that the software is running correctly.

@dash. The spikes in rpm (particularly low rpm) are because of the signal threshold setting (most likely). I have change the way this works because for a coil magnet setup you will have on part of the signal (say it's high side) will show two bumps each time the mag passes, but the the low side will have just one bump. On the two bunp side, the second bump gets bigger as the rpm go down and, at low rpm, this second bump gets picked up as a real magnet pass. Because it is so close to the real signal, the time between them is small so the software thinks that the rpm shot up (thus the spike). So run the dyno by hand, figure out whether your low side or high side has the single bump and then set the threshold. Set is below 125 or so to pick up a clean single bump on the low side, or above 132 or so if your clean single bump is on the high side.

To adjust torque scale, select the torque check box, select N.m as the units and the change the number in the torque Y axis to the level you want. The number in this box is always in N.m even if you select other units to display your torque.

Remember, the torque and power numbers may be very noisy based on the construction of your dyno (mine are all over the place).

[-dash-]

Ok had V.3 on laptop now not crashing as much but there is alot of lag:(

and been playing away with it on desktop trying to get rid of them naughty littel spikes just cant seam to get rid of them tweeking linein and threshold.

@Damo not sure if im being realy stupid or not but cant see low and high signal threshold there is just signal threshold that moves the high side of the bar.

and not getting 2 bumps on either side might just be misunderstanding on my part tho

Doh forgot to ask do you think it would matter how long the cable is that goes from pc/laptop to the dyno?

Edited July 3, 2010 by -dash-

[DamoRC]

@dash

When you say it's not crashing as much - when it crashes, what error is it posting?

If you have significant lag the data will be no good.

There is only one signal threshold now. Set it above the midline level (which is 127) to use the high side signals or below the midline to use the low side signals. The software won't let you set it too close to the midline)

Your signal is very clean and does not have the extra bump that I get so you can use either high ow low side signal.

On the picture you posted, this looks fine (very clean signal) but I am wondering how you captured the image because if this is a screen capture, there should not be a horizontal red line running straight through the signal. If you used a camera then this is fine because the image has probably got multiple samples in it (this signal box should refresh about 15 - 20 times per second.

The cable length should not matter except (i) the signal intensity will be lower due to line losses and (ii) you make pick up some noise on the line - but that does not seem to be the case given the image you posted.

Do you get lag on the laptop all the time? If you see lag them close the software and start it again does it get any better?

[-dash-]

no just quick image done in paint just to give clear image of what i am getting . could never get good pic with the cam

ahhhh im with ya now on the threshold thing.

and think worked out why not getting 3 waves coil i made does not have a metal core its just plastic thats the only thing i can think of that would add the 3RD wave.

yeah lag on laptop non stop

might get my old old desktop pc running XP and see if that lags find out if its just the laptop being low on power to process this type of info

[DamoRC]

The metal core thing might be the difference with respect to your signal shape. Also the alignment of coil and magnet (I have noticed that if I point my coil at the magnet at slight angles, the second bump diminishes significantly). So I think you are good there. The big thing is the lag. I don't think that it is lack of processing power on your laptop. My laptop is a couple of years old running XP and it runs fine (but occasionally I get the lag problem which is solved but starting and stopping again). Rather, I think it's the complexity of the audio setups in the newer OS versions or the fact that the program is not getting high priority from the OS.

I am looking for solutions to this that could be dealt with in the software.

[-dash-]

right gona take quick look see if laptop came with Win 7 disc if it did i will uninstall and reboot with XP or Vista sure got some copys kicking about and see if that does anything

[Fr0sty]

right gona take quick look see if laptop came with Win 7 disc if it did i will uninstall and reboot with XP or Vista sure got some copys kicking about and see if that does anything

@dash if your going to reinstall opt for XP its the most stable and compatible

Im going to have a good play tomorrow got some spare time and i am still buzzing with my 4th in the TT-01 A Final tonight

[-dash-]

yeah gona put xp on if can find the disc vista just eats up ram for no good reason and the stupid admin thing.

Put all driver disc etc somewhere safe but can recall where lol

nice work on the 4TH place bud

[DamoRC]

I'm sorry that the software is putting you to so much trouble. I have been trying to research ways to get rid of the lag but to no avail. I did read that updated drivers for you audio might help so I updated the drivers on my Vista desktop but it did not help.

[-dash-]

lol whats life without few tests here and there dont sweat it bud.

i dont mind what OS my stuff runs and no big deal to change it.... well when find the discs lol got my mate on the case now as he probaly got few XP and vistas kicking about.

really should give him the laptop and program really he prob go in 3 secs here your problem.

he is a programmer tec support fixxer person. fixed my desktop when got it as GFX cards where playing up

[DamoRC]

I put up version 3.1. This fixes the bug where a high or noisy signal crashes the software. I haven't changed anything else but, strangely, after running it a few times on Vista, it seems to be performing without any lag now (at least on my machine).

DamoRC

[DamoRC]

Actually I tell a lie. I was using the currently posted version last night and I noticed that the signal window seems to flicker a lot in this version. When I checked the code, I had modified it to use smaller sound buffers to reduce latency. Although I had tried this before and it didn't work, it may be helping a little.

Still seems to be working fine on my Vista desktop with no lag.

DamoRC

[Iceyankee-Tsi]

Greetings from Puerto Rico my friends!!!

I have been following this thread for a while and I am in the process of building my own dyno. At this time I am in the final stages of the build. I am making the casing for the roller assembly. I will probably use wood like Dash made his. I really need to acknowledge Damo Rc for this great piece of software and Dash for some outstanding craftmanship. A true piece of art. I have tested the software in a Windows XP computer with superb results. However, I cannot say the same for my Windows Vista computer. Even though it doesn't crash that much, magnetic pulse readings seem to be extremely sensitive and somewhat inaccurate. I understand it can mostly be related to the mic software been used in my laptop(Sony VAIO VGN-FZ180E).

I have been able to test the software in my XP computer and I find it extremely useful. Once I have completed the dyno, I will be using it on my Xray NT1 1/10 RC nitro touring car. I will be putting to the test my Picco Evo 3 red dot nitro engine.

In order to see If I am doing okay, I will like to ask a little question. After calculating my ideal roller mass, it came up to 1968.5 grams. However, after weighting my rollers it came to a total of 2297 grams. Do you think that additional 328.5 grams of mass may have a big impact in the software's calculations?

Thanks a lot and keep it coming!!!

Andres

[DamoRC]

@Iceyankee-Tsi,

The extra 328 grams will not affect the calculations. The extra weight will just put the car under a little additional stress and it will take a little longer to get to top speed. I don't think that this will be an issue.

Looking forward to seeing your build.