V3.0 mainboard trouble-shooting

Ask questions about DIY board assembly issues, fault finding and testing. (Covers all Megasquirt versions and board.)

Moderators: jsmcortina, muythaibxr

billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

V3.0 mainboard trouble-shooting

Post by billr »

This board was sent to me by the OP in this thread:

www.msextra.com/forums/viewtopic.php?f=131&t=57839

In a nut-shell, there was no CKP signal unless the MS was powered by a totally separate battery. I am trying to fix this for that OP and have a good scope as well as good V3.0 board for comparison, if needed, but will need (lots?) of help here.

The board looks good, there are a few suspicious solder connections, but no signs of anything being burnt are damaged.\

Since one of the symptoms was losing com with TS, I checked out the U5 operation right away, testing voltage at Vref. With 12.6V "raw in" (DB37-28) Vref is 5.06V. If I load the 5V (at the proto +/- pads) with 10.5 ohms, Vref drops to 4.77V. Same results if "raw in" is 8.6V. There is no DB attached and no other extra loads on the MB except a MAPDaddy in place of the standard MAP sensor, a 1K pull-up for the Hall CKP sensor, and a simple 2N2222-based driver off of JS7, looks like a 1K pull-up there also. Is 4.77V too low for that 5V to dip under load (about 450mA)?

Next, I have been focusing on the "tachin" circuitry. One thing that has me really confused is that I am seeing a square-wave (SQW) of 0-3.92V at both the VRout pad and the 40-pin DB socket (#14, IRQ1). However, the signal at both U7-7 and the board via connected directly to it is a SQW of 1.2-3.92V. Both the U7 VCC and ground seem solid, about 100mV P-P of hash on either. According to the schematic VRout is connected directly to U7-7, so how can there be that 1.2V difference on the "low"??? I checked probing both the U7-7 pin, and its pad on the solder-side, same results both places.

That's enough to chew on for now. I'm going slow here, trying to "do no harm" as I learn and try to help this guy out. Any help here will be appreciated!

Edit: 195K ohms between VRout pad and U7 pin 7 pad (or the nearby via hole). The trace from U7 to VRout is on an internal layer, no way for me to check it further?
Last edited by billr on Sat Apr 25, 2015 6:05 pm, edited 1 time in total.
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

Okay, I'm past part of my stupidity. I should have been looking at U7 pin1, that's the one connected to VRout! U7-7 is the VRout_inverted...
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

What's the proper size pin to use for a "drag test" of the 40-pin DB/CPU sockets? I tried using a round pin .0175" diameter and most had some grip, but some were loose. That .0175" seems kind of small to me, but it is a machined and gold-plated one from an old Augat component holder, I would expect it is a quality part and a legit size.
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: V3.0 mainboard trouble-shooting

Post by racingmini_mtl »

billr wrote:What's the proper size pin to use for a "drag test" of the 40-pin DB/CPU sockets? I tried using a round pin .0175" diameter and most had some grip, but some were loose. That .0175" seems kind of small to me, but it is a machined and gold-plated one from an old Augat component holder, I would expect it is a quality part and a legit size.
That seems like the right size: http://assmann.us/specs/arxx-hxl-tt-r.pdf (see the mechanical section on the right).

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

Thanks, I'll have to think about that a bit. There are several that seem like they have no grip at all.
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

This board used to have an MS2 DB on it, what is the pin size on an MS2?
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: V3.0 mainboard trouble-shooting

Post by racingmini_mtl »

billr wrote:what is the pin size on an MS2?
0.018"
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

How important is pin 1 on that 40-pin socket? I did some better "drag testing" today, sockets 1, 3, 17, 19, & 21 all have poor retention. Socket 1 is, by far, the worst. My "test pin" weighs less than 1/2 gram and falls out of that socket if the board is turned upside-down. I'm not expecting the 50 gram specified for a new socket, of course, but there is virtually nothing here...

I checked U5 regulation on my (good) V3.0, it is comparable to the one I'm working on; so I don't think that is an issue.
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

With my MS3 on it, I have run it up past 7000 rpm with both 12.8V and then 8V for raw power. It goes past 7K (8-9K?) with no problems on 12.8V, on 8V it starts losing rpm/sync at 7000. That may be because of my Hall sensor, made from an Allegro chip and powered off the +12. Worse yet, the air gap is about 2mm to a small-toothed wheel. I guess I have to try it on my V3.0 and see if it will go past 7K on 8V...
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

Do the DIY Hall CKP sensors require an external pull-up?
ChevytoyR1
Master MS/Extra'er
Posts: 443
Joined: Mon Oct 27, 2008 12:15 am
Location: Los Angeles, CA

Re: V3.0 mainboard trouble-shooting

Post by ChevytoyR1 »

No the DIY doesnt require a pull up
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

ChevytoyR1, this is from your original thread about this problem:

"Both ms3 on the stim work the same, except problem ms3 at really high rpm shows endless amount of lost sync errors."

I have my JimStim assembled now and am ready to duplicate your testing, so want clarification on a few things

1) What rpm did this bad board crap out at? Did the good one ever get sync loss; if so, at what rpm? If not, how high an rpm did you go to?

2) You had the Stim set for 60-2 and no inverting of signal (DIP switch #4 on only)?

3) The tach jumper on the Stim is set for Hall?

4) Which pull-up are you using on the Stim, 5V or 12V?

5) Any other jumpers/wires on the Stim that I should duplicate?

6) What input voltage to the Stim were you using?
ChevytoyR1
Master MS/Extra'er
Posts: 443
Joined: Mon Oct 27, 2008 12:15 am
Location: Los Angeles, CA

Re: V3.0 mainboard trouble-shooting

Post by ChevytoyR1 »

Image


The sync loss happens when the rpm on my stim is fully turned 16k rpm on the stim. All other boards i have dont loose sync at any rpm even at 16k
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

OK, pull-up to 12V on both primary and secondary. What about the power to the stim, what voltage do you use for that (measured)?
ChevytoyR1
Master MS/Extra'er
Posts: 443
Joined: Mon Oct 27, 2008 12:15 am
Location: Los Angeles, CA

Re: V3.0 mainboard trouble-shooting

Post by ChevytoyR1 »

If i remember right 12.10 volts
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

I'll set my supply to that, but I don't think it matters. In my previous experiments there seemed to be no difference whether using 6.5 or 12.5V for power.

I'm playing with my board right now, it loses sync at about 10K rpm with the simple "R56 six turns CW" pot adjustment. If I fiddle with R56, I can sneak the rpm towards 14K, but sync is gone for sure past that.

Both your board and mine have an LM2904 for U7. How about the board that worked OK for you, does that have an LM2904, or the MC34072AP? With the scope I can see that the JimStim output and the signals to U7A (pins 2 & 3) are all stable as I go through the sync/no-sync rpm zone. However, the output of U7A (VRout or U7-pin1) definitely does not stay stable. At lower rpms that VRout looks like a nice square-wave (4.4V peak). As rpm increases it becomes a trapezoid (still 4.4V P), and the top of the trapezoid shortens until the wave it a triangle pulse at 9K rpm. Increasing the rpm past that results in the VRout (triangle-wave) reducing in amplitude and pulse-width until sync-loss starts with the wave at 3.25 peak. So, I'm wondering if the LM2904 isn't quite fast enough for the higher rpm. Obviously, these findings should have nothing to do with your original problem of "no CKP signal while cranking", but I'm trying to understand all I can about that tachin circuit on the mainboard.
ChevytoyR1
Master MS/Extra'er
Posts: 443
Joined: Mon Oct 27, 2008 12:15 am
Location: Los Angeles, CA

Re: V3.0 mainboard trouble-shooting

Post by ChevytoyR1 »

Im not sure what u7 the new board has since its in the car now
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

I looked at the datasheets for those two versions of U7, the slew-rate for the MC device is over 40X faster than the LM; 13V/usec vs .3V/usec. Yeah, those are "typical" values, and under power/drive conditions not quite the same as in MS (maybe not even each other's test), but still... 40X. I checked slew rate for my LM2904 with the scope, got .269V/usec, quite close to the datasheet value. I am concluding that a 10K rpm limit is about normal with the LM2904 and a 60-tooth crank wheel. That high-rpm test using the stim was an irrelevant "red herring".

So, we are back to the issue of why there was no CKP signal (or sync, at least) unless the MS was powered by a separate battery. I'm going to move back to testing with the "bad" V3.0 board again, and am open to suggestions on what to look for.

I'll ask again... what are the chances that the loose socket, at pin 1 of the 40-pin CPU socket, is causing a "tachin" problem? I need some help here from real electronics experts!
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

*bump*

I'll try my question in a different way:

That pin 1 of the CPU socket is labelled "Vsyn" and "VDDA"(?) on the schematic. What do those terms mean, how does the MS3 board use that voltage? Since it comes from the 5V supply through an inductor for filtering, like Vref at pin 31, I'm thinking that Vsyn is important. However, if it would have no influence in detecting the VRout/IRQ1, then I would like to be able to ignore that loose socket.
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: V3.0 mainboard trouble-shooting

Post by billr »

Alas, I'm getting no bites on this one; I'm resigned to giving up and returning the board to its owner.

ChevytoyR1, I will send it out later this week, will advise when it is in the mail. Don't worry about the $6 to ship, I feel like I owe you that because I have failed.

I still feel that CPU socket pin 1 is bad, and could be the problem. I would fix that by putting another socket on top of it and soldering it to the board socket, not try to remove the existing 40-pin board socket. That might "ruin" the board for use with MS1 or MS2, as the added height might not fit in the case, but would not matter at all with an MS3 since there is a stack of sockets there normally. I would also re-heat pins 1-3 of U7, especially pin2. I do not think those solder joints are a problem, but they don't look as perfect as I would like, so would touch them up to be sure.
Post Reply