Had MJ fitted less than a week. Noticed drop in performance and found that there's no SAW output from MJ. There are no visible dry joints or similar so concluding that a component has failed.
Please help me if you can as I'm struggling
Keith.
EDIS wont advance - no SAW output
Moderators: JeffC, rdoherty, stieg, brentp
EDIS wont advance - no SAW output
Hi,
Had MJ fitted less than a week. Noticed drop in performance and found that there's no SAW output from MJ. There are no visible dry joints or similar so concluding that a component has failed.
Please help me if you can as I'm struggling
Keith.
Had MJ fitted less than a week. Noticed drop in performance and found that there's no SAW output from MJ. There are no visible dry joints or similar so concluding that a component has failed.
Please help me if you can as I'm struggling
Keith.
Keith S.
I take it you're getting an
I take it you're getting an RPM reading on the software? If so, check the following:
- signal on tach_out
- trace the signal back to the processor- check for a signal on pin 13 of the processor.
- signal on tach_out
- trace the signal back to the processor- check for a signal on pin 13 of the processor.
Painstaking fault finding
Painstaking fault finding I'm afraid....
- can you change the map with the configurator?
- does it show a realtime map?
- if you set the configurable outputs to, say, 3000rpm, do they happen?
- is the PIP signal definitely getting to the processor?
- is the SAW signal coming out of the processor?
- is +12v definitely getting to the board?
- is 'earth' definitely getting to the board?
- is +5v definitely coming out of the regulator?
- is +5v definitely at the supply pins of all the chips?
- is the processor oscillator working (need an oscilloscope for this!)
- is there any smoke (remembering all electronic devices contain smoke and when it gets out, you're generally in big trouble!)?
HTH and report back
Martin
- can you change the map with the configurator?
- does it show a realtime map?
- if you set the configurable outputs to, say, 3000rpm, do they happen?
- is the PIP signal definitely getting to the processor?
- is the SAW signal coming out of the processor?
- is +12v definitely getting to the board?
- is 'earth' definitely getting to the board?
- is +5v definitely coming out of the regulator?
- is +5v definitely at the supply pins of all the chips?
- is the processor oscillator working (need an oscilloscope for this!)
- is there any smoke (remembering all electronic devices contain smoke and when it gets out, you're generally in big trouble!)?
HTH and report back
Martin
Have tried some of that...
Thanks Martin and Brent for your suggestions, I'm competing off-road this weekend so re-fitted dizzy for now and will resume diagnostics Monday evening.
However I've already checked....
Software comms ok
Realtime display ok
PIP ok (at Molex)
No SAW (at Molex or Board or nearest resistor)
12v, 5v and Earth OK at all diagnostic points on board (shown on the assembly webpages)
Non-smoker!!
If anyother clues/pointers spring to mind please let me know.
Keith S.
However I've already checked....
Software comms ok
Realtime display ok
PIP ok (at Molex)
No SAW (at Molex or Board or nearest resistor)
12v, 5v and Earth OK at all diagnostic points on board (shown on the assembly webpages)
Non-smoker!!
If anyother clues/pointers spring to mind please let me know.
Keith S.
Keith S.
Try the other side of the
Try the other side of the resistor- the SAW output circuit is super simple: it goes Pin 13 from processor -> 100 ohm resistor -> molex connector. it also drives the tach output via the 2n2222 transistor, so that's why you would want to check both outputs.
Good luck.
Good luck.
Pin 13 reading 0.1v - 0.25v
Brent,
On Pin 13 getting 0.1v at tickover rising to 0.25 at mid-revs. This seems low to me. Also checked with meter from Pin 13 to SAW on Molex - measured 100ohm as I would expect.
Does this mean I've got a duff Processor?
(odd that I'm getting tach output though!)
Keith S.
On Pin 13 getting 0.1v at tickover rising to 0.25 at mid-revs. This seems low to me. Also checked with meter from Pin 13 to SAW on Molex - measured 100ohm as I would expect.
Does this mean I've got a duff Processor?
(odd that I'm getting tach output though!)
Keith S.
Keith S.
Still no Advance!!
I'm still plugging away but getting the feeling I'm just digging a big hole!
Summary:
No advance seen with strobe
Software ok and seeing runtime display
Tach output works (although stops at 2700rpm due to pulse width I guess - the least of my problems at the moment)
Seeing small a.c. voltage on SAW output which varies with rpm (also see the same readings at the EDIS SAW input terminal.
Have since taken the MJLJ to a friend who not only has a Scope, but knows how to use it AND has built a couple of Megasquirt units recently.
Even better that that he also has an "engine simulator" to generate PIP signals.
Hooked-up the MJ and did measurements on the SAW output and observed the SAW Pulse Widths changing with rpm changes (i.e. approx 1 sec when 10deg and approx 500mS when 40deg advance).
What I didn't note was the voltage - was a "few volts" I think.
From this we concluded that MJ was working correctly. I then hurried back home with a smile on my face and spare EDIS at my side.
Swapped the EDIS and no change
Tried yet another EDIS and still no advance
Checked-out PIP and SAW wiring again with meter but that was ok also.
Any suggestions where I go from here? (apart from the pub of course!)
The only nagging doubt in my mind is that the voltage of the SAW pulse from MJ is too low for EDIS to recognise - does anyone know what this should be when measured with a Scope?
Also, should there be a minimum distance between EDIS and Coils to prevent interference? Mine are around 6" apart could this be too close? I have used screened cable all earthed at same point.
Keith
Summary:
No advance seen with strobe
Software ok and seeing runtime display
Tach output works (although stops at 2700rpm due to pulse width I guess - the least of my problems at the moment)
Seeing small a.c. voltage on SAW output which varies with rpm (also see the same readings at the EDIS SAW input terminal.
Have since taken the MJLJ to a friend who not only has a Scope, but knows how to use it AND has built a couple of Megasquirt units recently.
Even better that that he also has an "engine simulator" to generate PIP signals.
Hooked-up the MJ and did measurements on the SAW output and observed the SAW Pulse Widths changing with rpm changes (i.e. approx 1 sec when 10deg and approx 500mS when 40deg advance).
What I didn't note was the voltage - was a "few volts" I think.
From this we concluded that MJ was working correctly. I then hurried back home with a smile on my face and spare EDIS at my side.
Swapped the EDIS and no change
Tried yet another EDIS and still no advance
Checked-out PIP and SAW wiring again with meter but that was ok also.
Any suggestions where I go from here? (apart from the pub of course!)
The only nagging doubt in my mind is that the voltage of the SAW pulse from MJ is too low for EDIS to recognise - does anyone know what this should be when measured with a Scope?
Also, should there be a minimum distance between EDIS and Coils to prevent interference? Mine are around 6" apart could this be too close? I have used screened cable all earthed at same point.
Keith
Keith S.
Surely not 1 sec and 500 msec?.
The formula for calulating the SAW pulsewidth (http://picasso.org/mjlj/?q=node/2) is:
Pulsewidth (usec) = 1536 - (25.6 x (advance in degrees))
Range: 64 - 1792us (57.5° BTDC to 10° ATDC)
Resolution: 4us (5/32nd of a degree)
So......
10 deg advance = 1536 - 256 = 1280us = 1.28ms
40 deg advance = 1536 - 1024 = 512us = 0.512ms
...so I presume you mean you observed about 1ms and 500us rather than what you say?
Can you get your friend to put his scope on your MJLJ whilst it is running on the car to see the voltage of the SAW pulse?
I'll put my scope on my unit (tomorrow night) and report back..
Martin
Pulsewidth (usec) = 1536 - (25.6 x (advance in degrees))
Range: 64 - 1792us (57.5° BTDC to 10° ATDC)
Resolution: 4us (5/32nd of a degree)
So......
10 deg advance = 1536 - 256 = 1280us = 1.28ms
40 deg advance = 1536 - 1024 = 512us = 0.512ms
...so I presume you mean you observed about 1ms and 500us rather than what you say?
Can you get your friend to put his scope on your MJLJ whilst it is running on the car to see the voltage of the SAW pulse?
I'll put my scope on my unit (tomorrow night) and report back..
Martin
-
Oliver Sedlacek
- Posts: 68
- Joined: Mon Sep 13, 2004 8:45 am
- Location: The Chalfonts
- Contact:
How about
The wiring. In my experience the cheapest, simplest bits cause the most problems. As you've changed the EDIS and megajolt I suspect the connector s and wiring.
OK, I did it tonight!
Not very accurate measurements, but may be useful.
At idle (about 1100rpm) my PIP waveform is close to a square wave, 10ms on then 10ms off, min 0v max 12v. I can't quite work out why 1100rpm equates to a 20ms period PIP wave, but no doubt someone will tell me!
The SAW waveform goes from 0v to +5v when the PIP waveform goes from 12v to 0v, and stays at 5v for about .9ms (900us) then returns to 0v. I think my map is currently 15 deg advance at idle, so that computes about right.
When I rev the engine, the frequency of the PIP wave increases, still min 0v, max 12v and the SAW wave stays at 5v for less time (around 400us) at a few thousand rpm. Just as predicted by the formula, and as required.
Looking at the circuit diagram confirms that the SAW wave output should be 0v min, 5v max.
These timings and voltage levels are incredibly similar to the text description above the waveform shown in the second diagram on http://www.bgsoflex.com/mjl/mjl_edis_summary.html
You really need to get the scope onto the car to find out what's going on - using an analogue or digital voltmeter on an AC (or DC) scale to measure square waves (PIP at idle) or pulse waveforms (SAW) is like trying to examine the PCB for soldering errors through misted up glasses in the dark
At idle (about 1100rpm) my PIP waveform is close to a square wave, 10ms on then 10ms off, min 0v max 12v. I can't quite work out why 1100rpm equates to a 20ms period PIP wave, but no doubt someone will tell me!
The SAW waveform goes from 0v to +5v when the PIP waveform goes from 12v to 0v, and stays at 5v for about .9ms (900us) then returns to 0v. I think my map is currently 15 deg advance at idle, so that computes about right.
When I rev the engine, the frequency of the PIP wave increases, still min 0v, max 12v and the SAW wave stays at 5v for less time (around 400us) at a few thousand rpm. Just as predicted by the formula, and as required.
Looking at the circuit diagram confirms that the SAW wave output should be 0v min, 5v max.
These timings and voltage levels are incredibly similar to the text description above the waveform shown in the second diagram on http://www.bgsoflex.com/mjl/mjl_edis_summary.html
You really need to get the scope onto the car to find out what's going on - using an analogue or digital voltmeter on an AC (or DC) scale to measure square waves (PIP at idle) or pulse waveforms (SAW) is like trying to examine the PCB for soldering errors through misted up glasses in the dark
Sounds about right
Thank you for your suggestions. Those scope readings sound very similar to what we observed so helps confirm that MJ is probably working correctly.
I plan to confirm that my EDIS units are ok by trying them in friends car this weekend. Have already buzzed-out the wiring and that's ok so will look more closely at the screening.
If necessary we'll ge the Scope hooked-up to the car, that'll tell me for sure, but can't do this until next week.
Keith S.
I plan to confirm that my EDIS units are ok by trying them in friends car this weekend. Have already buzzed-out the wiring and that's ok so will look more closely at the screening.
If necessary we'll ge the Scope hooked-up to the car, that'll tell me for sure, but can't do this until next week.
Keith S.
Keith S.
Confession time.....
Well I've finally found the problem that's been bugging me for ages and YES it was something very simple in the end.
In my infinite wisdom, for some reason that will now give me global international embarassment, I goofed on the wiring, well not exactly the wiring but the SCREENING!
I'd conected connected both screens to the correct terminal on the EDIS but assuming that the screen was going to earth (which I now know it doesn't), I also earthed the screen at the MJ end on molex plug. So the simple golden rule - DO NOT EARTH THE SCREEN's
Anyway looking on the positive side I have now learned a fair amount about the MJ and I have 2 spare EDIS units (that I'll probably never need!!)
Thank you to everyone who helped with suggestions - this forum works really well. And you never know someone maymake the same stupid mistake as me, in which case this posting may hlp get themselves sorted without public humiliation!!
Now I can get on with some mapping.
Regards,
Keith.
In my infinite wisdom, for some reason that will now give me global international embarassment, I goofed on the wiring, well not exactly the wiring but the SCREENING!
I'd conected connected both screens to the correct terminal on the EDIS but assuming that the screen was going to earth (which I now know it doesn't), I also earthed the screen at the MJ end on molex plug. So the simple golden rule - DO NOT EARTH THE SCREEN's
Anyway looking on the positive side I have now learned a fair amount about the MJ and I have 2 spare EDIS units (that I'll probably never need!!)
Thank you to everyone who helped with suggestions - this forum works really well. And you never know someone maymake the same stupid mistake as me, in which case this posting may hlp get themselves sorted without public humiliation!!
Now I can get on with some mapping.
Regards,
Keith.
Keith S.