M2-ATX(-HV) 160 watt Intelligent DC-DC PSU - Page 8

alpax · 10-21-2009, 11:10 PM

Quote: Originally Posted by tbird2340

Doesn't like the "shape"? What does that mean?

The shape of the line on a graph of voltage, for example. If the ignition signal does not slowly and constantly fading when turned off, but fluctuates up and down a bit with reducing amplitude. But again - it's just a guess.

Quote:

I don't get it.. If the hot line is getting a constant 12V then how can it totally cut power when the IGN line loses power?

There is some switch logic and a microcontroller on that board, which gets ignition signal, interprets it and controls the output power. I guess it does not interpret the fluctuating ignition signal properly - maybe due to some error in a microprogram, and cuts the power at a wrong moment.

Quote:

And it doesn't power back on after that like some have said.. It totally cuts power.

Well, it could be something else then. But still, sounds similar to other reports.

Mickz · 10-22-2009, 12:59 AM

Just a thought as to with what MAY be happening on the 12v line. It sounds like you are getting a pulse on the 12v line. You cannot see this with a meter, you need an oscilloscope.

Microprocessors and electronics are very sensitive to a fast rising short pulse.

This pulse may be injected directly into or coupled into the 12 volt line or other cable going to the supply. The fact that a filter fixed the problem indicates that this is a very fast rising short spike on the line.

It is obviously being coupled into the ATX supply and due to the nature of this pulse it is getting past the filtering circuitry and tripping the controller.

tbird2340 · 10-22-2009, 08:03 AM

Quote: Originally Posted by Mickz

Just a thought as to with what MAY be happening on the 12v line. It sounds like you are getting a pulse on the 12v line. You cannot see this with a meter, you need an oscilloscope.

Microprocessors and electronics are very sensitive to a fast rising short pulse.

This pulse may be injected directly into or coupled into the 12 volt line or other cable going to the supply. The fact that a filter fixed the problem indicates that this is a very fast rising short spike on the line.

It is obviously being coupled into the ATX supply and due to the nature of this pulse it is getting past the filtering circuitry and tripping the controller.

If that was the case you would think I would get that during crank, not simply turning the vehicle off, no?

It survives fine past the crank and works great up until turning the IGN off..

First thing I'm going to try is connecting the switched wire to an accessory wire instead of the ignition wire.. If that don't work.. How do I fix it? I have that same liliput thing that was mentioned but where do I hook it up?

Thanks

Mickz · 10-22-2009, 07:31 PM

Quote: Originally Posted by tbird2340

If that was the case you would think I would get that during crank, not simply turning the vehicle off, no?

It survives fine past the crank and works great up until turning the IGN off..
Thanks

Well that’s the $@#!* thing with electronics, starting pulls 100+ amps and drops the voltage, however, because of the way it’s connected, the battery sinks this pretty effectively and your left to contend mostly with a low voltage running condition, which the PSU seems to handle ok.

If it is indeed a pulse causing the problem when shutting down the ignition then it could be generated by back EMF from some other inductive device in the car.

I assume that that PSU is not worried about its control line going hi/lo a few times in succession as the ignition is turned off. This often occurs because of switch bounce, which is why all good electronic gear has de-bounce circuitry on all control lines.

It’s a shame you don’t know someone with CRO (and who also know how to use it correctly) as it would take a few minutes to either eliminate or pinpoint the problem.

Not being familiar with that DC-DC PSU I’m sorry that I can’t be much help.

tbird2340 · 10-22-2009, 08:51 PM

Well I was going to try hooking up to an accessory line but my vehicle doesn't have a regular accessory line it's a "mux" line so it won't work..

What can I do now!? How can I get this to work!?

Mickz · 10-22-2009, 09:58 PM

Quote: Originally Posted by tbird2340

it's a "mux" line so it won't work.. What can I do now!? How can I get this to work!?

I don’t want to lead you down the garden path here but if I was having this problem and I had eliminated the rest of the wiring then I would run a small temporary wire (run it outside the car and through the window for the test) directly from the battery to the control wire and see what happens when the car is turned off with that line connected.

If the PC runs ok and does not immediately power down when the motor is shut down then you know for sure that it’s the control line that is the source of the problem. At Least you then have a place to start.

tbird2340 · 10-22-2009, 10:03 PM

Good idea.. I'll try that tomorrow.. Thanks for the help.

tbird2340 · 10-22-2009, 10:16 PM

Quote: Originally Posted by Mickz

I don’t want to lead you down the garden path here but if I was having this problem and I had eliminated the rest of the wiring then I would run a small temporary wire (run it outside the car and through the window for the test) directly from the battery to the control wire and see what happens when the car is turned off with that line connected.

If the PC runs ok and does not immediately power down when the motor is shut down then you know for sure that it’s the control line that is the source of the problem. At Least you then have a place to start.

On second thought.. How exactly do you mean to do this? Should I start the vehicle and then connect another 12v constant directly from the battery to the switched pin of the M2, wait for it to power on completely, and then turn the vehicle off?

Mickz · 10-22-2009, 11:04 PM

Yes, it's not the ideal way to do it because it may be a combination of the power down signal and the engine shutdown, but the idea is to see if the cause of the sudden PC switch off is that control line or some other connection to the PC/PSU. Then a more correct test can be carried out.

tbird2340 · 10-23-2009, 08:17 PM

Quote: Originally Posted by WuNgUn

Without shutting of the IGN, remove the wire from the IGN leading to the M2 and see what happends...
Also, what do you mean the 'ground is to ground'? It's attached to the chassis of the car?

Ok, I started the car with the IGN wire disconnected.. With the car running I then applied the IGN wire to the M2.. It powered up correctly after 5 seconds.. I let it run for a minute or so and then removed the IGN wire from the M2..

It began the hibernation process.. While it was still in the process I turned off the car.. It didn't shut down and continued hibernating correctly..

So, it has to be something going on with the IGN wire only when the car is shut off because it works fine when I manually remove the IGN wire from the M2..

How can I fix this?

Thanks

Mickz · 10-23-2009, 08:25 PM

Just one more test, this time don’t set the PC to hibernate, just cut the ignition. Want to make sure the PC keeps running fine.

tbird2340 · 10-23-2009, 09:03 PM

Yea, it does..

Mickz · 10-23-2009, 09:29 PM

Ok, so you have now eliminated the rest of the installation and were down to something going on with the PSU control line.

Now there are elegant solutions to the problem BUT for now we want to keep it SIMPLE and almost NO COST until were 100% sure we have a fix.

So, with that in mind you could try a couple of quick fixes.

1. Try a quick filter on the control line.
Go to an electrical hobby store and buy a 47ohm ½ or 1 watt resistor and a 10uf 25 volt electrolytic cap. Place the resistor in line with the control line, mount it near the PSU, and then connect the (+) side of the cap from the PSU side of the resistor and the (–) side of the cap to a ground point on the PSU. See if it solves the problem.

2. Solution two.
Get a very small 12v relay. Power the relay coil from the IGN wire. Connect one of the relay “normally open” contacts to the PSU control line and the other to the “always on” 12v line going to the PSU. This “almost” completely isolates the IGN wire from the PSU control line.

Once again NO guarantee that this will work first time, but I am pretty sure we can get it to work one way or another. Again the point of the exercise is not to get technically cute, just to get it working first.

tbird2340 · 10-23-2009, 10:24 PM

I seen one member use the liliput egg / filter before.. Could I try that? If so, how would I wire that up?

Mickz · 10-23-2009, 10:32 PM

Yep, I mentioned that earlier, didn’t know if you had one though. That is most likely just a DC filter; connect it the same way, between the IGN and the PSU control line. He may have used his in a different location but we know that your problem appears to be coming down the control line. The filter I described will de-bounce the ignition line a lot more, but that should not be the problem anyway.

User Name		Remember Me?
Password