omega 24 v6
|
 posted on 27/10/08 at 09:43 PM |
|
|
Any electronic guru's want to help?
Having spent ALL DAY metering the megasquirt wiring loom on my car and finding NOTHING fundamentally wrong with it I wonder if any electronic
guru's would be willing to speak to me online or via U2U about a couple of niggles I'm having.
Basically electrics to me are no problem but electronics are a higher level and I'm finding it hard to comprehend where I may be going wrong or
making a mistake. I need advice on why it's better to ground one end of a screen cable rather than the other and how to go about getting rid of
electrical noise (If I have it I'm just not sure)
Getting desperate now as I'm ready for sva if I could only get to the stage that the engine would start first time every time the key is
turned.
I don't want to offend anyone but I'd rather if you were not a guru that you didn't reply with what may not be a logical answer.
If it looks wrong it probably is wrong.
|
|
|
|
|
BenB
|
| posted on 27/10/08 at 10:12 PM |
|
|
IF you connect both of them to earth you can get something called a ground loop.
My brain doesn't work beyond that point but if you look it up on the net you should find some answers...
|
|
|
omega 24 v6
|
| posted on 27/10/08 at 10:19 PM |
|
|
quote:
IF you connect both of them to earth you can get something called a ground loop.
Yes i know that but I want to know why grounding one end in particular would be better than the other. I had the vr screen grounded at the megasquirt
end and the car would not start or even offer to start. So I disconnected it and it fired pretty instantly and it's run up to temp 4 times like
this already but has trouble starting from cold/cranking.
Now today after metering the loom out I find that the screen is NOT grounded at the engine side at all (as I thought it was) so now there is no
screening grounded at all ( but the car has run like this only not anymore as there are IMHO perhaps other issues that need looking into).
I really need someone who understands electronics and electric noise to have a chat with and ask some questions.
If it looks wrong it probably is wrong.
|
|
|
MikeRJ
|
| posted on 27/10/08 at 10:44 PM |
|
|
If grounding the screen on the VR sensor is preventing the car starting, this would tend to suggest you have a short between one of the signal wires
and the screen.
|
|
|
omega 24 v6
|
| posted on 27/10/08 at 10:53 PM |
|
|
quote:
If grounding the screen on the VR sensor is preventing the car starting, this would tend to suggest you have a short between one of the signal wires
and the screen.
No they are defo clear. I believe the screen should be grounded at the engine end (not as I had it) I also thought I had it grounded at both ends
(which I don't) so I'll try grounding at the engine end tomorrow (if I get time)
If it looks wrong it probably is wrong.
|
|
|
tomblyth
|
| posted on 28/10/08 at 08:22 AM |
|
|
ground loop
getting rid of it
|
|
|
omega 24 v6
|
| posted on 28/10/08 at 09:20 AM |
|
|
Ah see now you are all asssuming I have ground loop problems. I reckon I have fuel pump interference problems that are affecting the spark moment. I
also nee to find out the pro's and cons of which/what items to use to filter out any noise. Where I should and should not group items of
equipment (relays and sensor wires. Whether I can feed certain items off of one fuse without risk of a feedback interference to more sensitive
electonic items etc etc.
This is why I wanted to speak to a Guru. I don't want to spend a whole weeks holidays trying to get this thing going by having various stabs in
the dark at what MAY BE a guesstimate of an answer.
But thanks for your input anyways so far.
If it looks wrong it probably is wrong.
|
|
|
Rob Lane
|
| posted on 30/10/08 at 09:35 AM |
|
|
This is one of the trickiest areas in electronics and vehicles.
Vehicles are a perfect noise generator and it is not always possible to turn off the noise source to achieve a result.
If you know anyone with an oscilloscope you can borrow then this is the best visual way to check noise input. Some is inevitable but the level of
noise is the key.
Grounding the shield at the engine is a way of 'extending' the shield protection along to the input. This is because it's the engine
end that is generating the noise signal, so extending the shield affords protection.
ECU's in their own right are also 'noisy' that's why design is important, along with signal isolation and seperation.
Personally I didn't get a clean crank signal till I completely isolated the wiring of the sensor and gave it a seperate route.
Battery main cables should be avoided and the worst culprit, alternator wiring should be given a wide berth.
FI pumps are noisy but if kept clear of any MS inputs should be OK. Not heard of this as a specific item to avoid. Again a scope on pump wiring would
show this up.
Bit general advice but hope it helps.
EDIT: Re-read your post again, I'm assuming you are having trouble with the crank signal, right?
It's important it is the correct way round, there is even a special section on this in MS manual.
[Edited on 30/10/08 by Rob Lane]
|
|
|
omega 24 v6
|
| posted on 2/11/08 at 08:19 PM |
|
|
quote:
EDIT: Re-read your post again, I'm assuming you are having trouble with the crank signal, right? It's important it is the correct way
round, there is even a special section on this in MS manual.
Ok guys thanks for all your help. For future references the problems were many.
Firstly the mixture was way rich (and i could smell it was).
Second the lc1 was causing a lot of interference on warm up. This became apparent as i had it as a plug in setup. Once while the engine ign was on I
plugged it in and the coil fired (fuel vapour ignited from the priming pulse) and flames out of the throttle bodies.
3rd I have a bottom of tank mounted fuel pump which was very noisy (in the hearing way) I was at the stage where I was only putting in a small amount
of fuel at a time. I did however get to the point where I new I'd have to fill the tank pre SVA to check for leeks. So 2 gallons were added and
the pump became much quieter.
4th I had fried a vb921 (dwell was to big and ign left on to long)
Now it seems all is fine and car is running a treat.
Problems as I see it were
Lc 1 was causing false triggers at startup
Fuel pressure may have been way to low due to not having enough fuel covering the pump.
Vb 921's replaced by BIP373 which in my opinion are MUCH MUCH better as they have a built in thermal cutout. So if you do overload them they
don't fry. I only discovered the dwell problem AFTER them heating up and cutting out twice. So I saved on having to get new vb's which are
now obsolete as well as grief wondering what was still wrong.
If it looks wrong it probably is wrong.
|
|
|
