JimStim 1.4 rpm burst

Building and using the Stimulator to test your Megasquirt

Moderators: jsmcortina, muythaibxr

Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

JimStim 1.4 rpm burst

Post by Masi_K »

My jimstim bursts rpm up to 1500-2500 rpm before settles down to what is turned from rpm fine pot. rpm coarse is okay. If i turn pot very very slowly then stim may rise rpm as it should. After it is settled, I can turn it how fast I want to. This isn't ignition mode related.
an Video tells more than my words above =) It's in divx3 and takes 1.2mb, no sound.
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: JimStim 1.4 rpm burst

Post by racingmini_mtl »

When you say this isn't ignition mode related, do you mean you tested it with different wheel pattern modes and it does the same thing in all of them?

This problem could be due to a dirty or faulty pot. What you could do, with the JimStim not powered, is to turn the pot fully from end to end a few times and it might be enough to clear some dirt. The best would be to use a pot cleaner spray. If this doesn't clear the problem then the pot may need to be changed.

Also, check the solders on the pot and on the CPU. If they are not very good, they may have a bad contact that makes intermittent connection and it would be more apparent when moving the pot fast because you put more force on it.

It might also be due to the some strange interaction with the firmware and configuration you're using. The fine RPM pot will only change the RPM by 512RPM from the lowest point to the highest point so if you're really seeing a jump from 1500-2500RPM then there is something strange happening. The exception is the dizzy mode when using a 4-cylinder setup where the range is actually doubled.

Also, you have to realized that the JimStim will change the interval between ignition pulses in a stepped manner so if the RPM is not really high and you change the pot position very quickly the code will see a very big RPM change between ignition pulses. I don't think this is the issue here since you say that you don't have problem when using the coarse RPM pot which makes this issue a lot more evident since each increment using this pot is 256 RPM.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

Yes it is not dip switch related. I just tried 3-5 dips up what ever modes they each represent. At every mode there is burst before pattern starts to roll. I measured a voltage at center pin of rpm pod against earth pins. There is no noise or spikes. Rpm starts to blink over 0.4V. Allso measured amount of the rapid voltage rise. 0.2V to 0.5V and there is burst. 0.5 to 1.8v and no burst. Few solder points at dip and cpu had nice lack layer on it. But they should be okay... 8)
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: JimStim 1.4 rpm burst

Post by racingmini_mtl »

There is nothing in the JimStim code that would cause that so the issue has to be in hardware (and I haven't heard of anyone else having this issue either which confirms that the code is not at fault). Either there is a problem with the CPU itself or one of the components or the soldering or any combination of these.

You could remove the CPU from the socket and reinstall it to make sure that there isn't any oxidation on the CPU pin and/or the socket. Also, you say that there is no noise but are you measuring with a meter or with a scope? A meter will not tell you much about the presence of noise. Since you get a very localized region where there is an issue this would seem to point to the pot being marginal at some point.

If you have a scope (and this isn't what you used) it would be a good idea to see if there is any noise present. Again from what you say, the pot seems to be the main suspect in this.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

I wish i have a scope :(
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: JimStim 1.4 rpm burst

Post by racingmini_mtl »

What you could try is to exchange the spare ADC pot with the fine RPM pot. You'll have to be careful desoldering the pots so that you don't overheat the board. Also before you do that, make sure that the spare pot doesn't show the same problem you see on the fine rpm pot.

Hopefully that will solve your problem. If it doesn't then the issue may be the CPU itself.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

I just done re soldering on those maybe bad joints, re seated cpu, tried to push dip switches little further and no help. Have to try pot change.
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

I wasn't able to swap the pots. What I can see, there is no shorts on board and joint's are good enough :| I think I was wrong about that the problem only exist fine rpm. It does exist at coarse too.. I Measured Volts from U1 pins. Pin 16 has negative voltage and that is fluctuating from 0.8 - 1.3V. When I just touch any ground points at board, it drops(or rises) to zero. I have no ideas.
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

User manual says that every pins expet 13, 15, 1 and 3 should read 0 or flote. According this, my board has error. :(
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: JimStim 1.4 rpm burst

Post by racingmini_mtl »

Check all solders carefully and clean the board with alcohol to make sure there is no flux residue.

If you can take good picture of the board on the top and bottom and post them, I'll have a look to see if I can find anything.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

Two pics asap:
Top view
Bottom view

Don't mind that barrel disortion what my camera produces. :roll:
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: JimStim 1.4 rpm burst

Post by racingmini_mtl »

It's a bit difficult to see but I don't really like how the bottom of the board around the fine RPM pot pads looks like. I understand that it may be because you removed or tried to remove the pot but first, you really need to clean the flux residue. Second, I would really make sure that the solder on the middle pin doesn't touch the pad or solder on the spare pot leg. Also, it may be due to reflections but it looks like the middle pin also has some solder on the other side that makes contact with the ground plane.

I don't see much on the coarse RPM pot but there is also some flux residue. This can cause all sorts of weird things when moisture gets into it.

So check those out and see if there is anything there.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

Ah, I didn't clean the board with alcohol at first place before assembly :( I should have. Maybe next time.. :RTFM:
I'll check those your findings.
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

I found no flaws after i cleaned board. I took two photos. Problem still persist.
Another angle from top
Bottom washed

Big pictrures this time. I only had cigarette lighter fuel for cleaning 8) . It took away residues, but surface is bit sticky.
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: JimStim 1.4 rpm burst

Post by racingmini_mtl »

Well, the board looks good. There may be the 2 ground pins on the RPM pots which solder doesn't look as good as the others but I don't think that's the issue. You could re-heat them to make sure they are correct.

Have you measured the voltage on pins 13 and 14 of the socket? If you measure the voltage at pin 13 (or 14) and move the coarse (or fine) RPM pot from one end to the other you should have a stable voltage from 0 to 3.3V. If not then there is a problem with the pots/board. If you have a nice stable voltage then the problem could be the CPU.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: JimStim 1.4 rpm burst

Post by racingmini_mtl »

I should add that you may want to re-try now with the cleaned board and it might have solved the issue.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

Both rpm pins at cpu has stable voltage rise from 0 - 3.3V. Tried with cpu back on board. No any difference. I tried to re heat all of those bad looking joints. How it could be cpu ? I have been very carfully with static electric etc. btw. that regulator warms up littlebit when power is on, but far from hot.
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: JimStim 1.4 rpm burst

Post by racingmini_mtl »

I had a look again at your video and there is nothing that shows it's not working correctly. Of course, is you turn the fine pot very slowly from when there is no RPM to when there is RPM, you can get what looks like a wrong RPM (and it is) but that's just because the value of the pot is just at the threshold between no rpm and the minimum rpm (60 RPM) so it oscillate between starting and stopping which looks like a much faster RPM. I'm not sure if that's clear but I can do exactly what I see in your video and I know my JimStim is functioning perfectly.

When I see the weird behaviour you show in your video, I can see the ADC value on my JimStim oscillating between 29 and 30 which is the threshold (on the fine RPM pot) between running and not running but only if the coarse RPM pot is turned off. And I can do a similar thing with the coarse RPM pot when it oscillate between 3 and 4. So when the selected RPM is above 60RPM (with whatever pot combination) then the RPM signal is really regular and smooth. By the way, anyone can have a look at the ADC values if they have the latest V2.0.3 firmware and the correct serial cable by using either Megatune or TunerStudio.

So, I don't think there is anything wrong with your JimStim and the RPM problem you're seeing is somewhere else. I would suspect either the MS circuit needs adjusting or you have a setup that's not correct.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Masi_K
Experienced MS/Extra'er
Posts: 199
Joined: Tue Aug 16, 2005 11:17 am
Location: Finland

Re: JimStim 1.4 rpm burst

Post by Masi_K »

Okay, I'm glad to hear that board is fine. I'll just have to turn pot(s) slowly at first quater of turn. :) Would any software delay for low adc values be workaround to this feature ? Just asking.
Volvo 244 gle, b19e msns. R.I.P.
Vauxhall Viva, Msns. "Finally working"
Volvo 945 Turbo -97, Ms II. "There can't be more resistant install than this"
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: JimStim 1.4 rpm burst

Post by racingmini_mtl »

Remember that this will only happen when you go from below 60RPM to above with either the coarse or fine RPM pot. And only when you stop just at the border of 2 ADC counts. Anywhere else you will have a potential jump between 2 RPM values: with the fine, it will oscillate by 1RPM but with the coarse, it will oscillate by 256RPM.

But you really have to try to do that or have dirty or faulty pots. The RPM will change for every 2 ADC counts on the fine RPM pot and every 4 counts on the coarse RPM pot. So you have to be on those specific counts to have the oscillation and you have to put the pot at the exact correct place between 2 counts to have the count oscillate. I have seen this very rarely and even trying to get it I have to be very patient and turn the pots extremely slowly by a very small amount.

So while this could probably be dealt with in software, I don't think it is a real issue for most people. And the JimStim is not intended as a lab spec equipment either and the CPU resources are scarce. It is possible that you have pots that are not perfect and are more susceptible to a bit of variation at just the wrong place and this is why you have been more affected. Another solution is to use the V2.0.3 firmware and the serial cable and control the RPM from the PC. This will guarantee a perfect RPM value without any unwanted variation.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Post Reply