where to get new//recon/used Bosch engine controller for Cavalier?

[Cross-posted to a few, hopefully relevant and/or useful, newsgroups]
Any pointers where to get a Bosch engine controller for a Vauxhall Cavalier (old-style, 1987, SRi 2.0)? Obviously I'll ring round Bosch dealers etc - just wondering if there are any cheaper outfits doing reconned units or repairs etc.
The Bosch number is 0 261 200 104.
Problem seems (according to our mobile tuning/repair man, who seems to have mucho clueo) to be that it's not driving the ignition coil, so prolly just a transistor gone. Is it get-attable/repairable?
Location is Reading.
tia
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

John, I can look at your controller for you
I doubt I can test the thing properly without at least a Haynes manual or something, but I can prolly locate an obvious fault
That's what I now seem to be good at
you can contact me via email (no anti spam measure there, or via the website (www.cetltd.com).
Unfortunately I will be out of the country from the 4th to the 14th, so if you can wait a week, I'll see what I can do
--
geoff

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
http://www.migweb.co.uk

Cavalier
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Try ATP - they'll check and supply re-cons for most makes.
http://www.atpelectronics.co.uk /
Remember reading that a large percentage of returns they get after supplying re-cons are found to be perfectly ok.
--
*Real women don't have hot flashes, they have power surges.

Dave Plowman snipped-for-privacy@argonet.co.uk London SW 12
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Mike Walker, the guy that started their repair service in the late 80's, told me the same. The failures were simply due to corrosion in the plug and socket connection. Only cleaning needed. Mike.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Mike G <mikgibbs at tiscali dot co dot uk> wrote:

As a sound engineer, I know only too well how often plugs and sockets can cause problems through 'dirt'. If I were a motor mechanic I'd just get the client to pay for a new mixer rather than cleaning them. ;-)
--
*I started out with nothing... and I still have most of it.

Dave Plowman snipped-for-privacy@argonet.co.uk London SW 12
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Thanks - ordred yesterday, received today (at about 11:59 and 240 seconds - 'am' delivery :-) fitted and running, for c. 120 compared with about 100 more that Cafco etc were quoting, also for next-day.
Could probably have repaired if it's just the coil-driver transistor that's gone, but I just needed to get it sorted.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

...
Well it _was_ running, but as soon as SWMBO tried it it didn't want to know. I tried and it started OK ... once, but refused to re-start after that.
I spoke to Mike (Walker?) at ATP who suggested several possible faults:
a) failing magnet on crankshaft position sensor - AIUI the ecu wants to see a couple of cycles of 58 pulses + 2 missing pulses from the toothed wheel on the crankshaft so it knows where the crankshaft is and how fast it's going before it'll fire the coil. If the magnet is losing strength the output from the coil can be too little at the low cranking speed for the ecu to be see the pulses it wants.
b) dodgy aluminium screening on some crankshaft position sensor cables which deteriorates allowing the cable to pick up noise corrupting the signal so the ecu doesn't get the 58 pulse sequences it needs to start firing the ignition
(Mike told me that the design of the ecu is such that it requires clean 58 pulse sequences before it will start firing, but once the engine is running it will keep firing even if the pulse sequence isn't clean: this is so that faults while running don't result in the ecu cutting the engine while the car's possibly doing 70 in the outside lane!)
c) dodgy contacts on the ignition switch, particularly a fault where the 'ignition' supply to the ecu cuts out when the key is turned to the starter position (I've had this fault on one Cavalier I've owned - possibly the current one, I forget which)
d) dodgy contacts on the dual fuel pump + ecu supply relay (2 relays in one box). This relay gets activated by the ignition key but is held on by a signal from the ecu itself so that rather than the ecu losing power immediately the ignition is switched off it can keep its own power on while it shuts itself down cleanly and then turn its own power off (like modern PCs which turn their power off after shutting down). Apparently (according to both Mike and my mobile mechanic) these relays were so notoriously unreliable that clued-up folks would keep a spare in the glove compartment! What can happen (according to Mike) is that the ecu-supply side relay's contacts can be a bit iffy and so the voltage at the ecu can be too low under starting conditions when the voltage is low anyway.
Does this all seem accurate?
If so I've got a few things to investigate:
1. the ignition switch and relay. I can measure the voltage between the switched +ve from the relay to the ecu and the battery +ve whilst cranking over the engine to see if I'm losing volts to the ecu. If I am I can jumper the power across to see if that makes it work.
2. The crankshaft position pickup. I know the existing one isn't completely broken because I can measure c. 500 ohms across the coil contacts, and I know the wiring to the ecu isn't completely open because attaching a spare pickup assembly (off an Astra, as it happens) and waving a screwdriver or other ferrous object around it makes the ecu respond by switching the fuel pump relay on. If I had a storage 'scope I could think about analysing the signal from the sensor but without that I can only think of swapping it for a new one (as I say, the 'spare' is off an Astra and is not the same as mine, I'm told). Unfortunately I gather that new ones cost c. 50 - 100 (ouch!). Anyone know of a cheaper source (or have one I could beg/borrow/steal in the Reading area?). I'm not hopeful of finding one at a scrappy as Cavaliers seem to be a bit thin on the ground in them these days.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Have you got a dwell meter? If it has an ordinary coil, a dwell meter should give a normal reading - showing the sensor is switching correctly.
--
*Why do psychics have to ask you for your name? *

Dave Plowman snipped-for-privacy@argonet.co.uk London SW 12
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

I haven't, but the coil is driven by the ecu - or rather should be driven by the ecu but it isn't driving it. I don't think it's a fault with the ecu (it's brand new, and the previous one was doing the same) and the coil's new too. That's why I'm investigating the pickup etc as described above.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Doesn't matter how the coil is driven - it still requires the field to be built up then broken to derive a spark. So a dwell meter will show if this is happening - even when cranking.

Well, it would eliminate the sensor, given that no pulses (or incorrect ones) mean no or incorrect dwell.
--
*Seen it all, done it all, can't remember most of it.

Dave Plowman snipped-for-privacy@argonet.co.uk London SW 12
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

correctly.
Err, not sure I'm with you, or maybe I described the problem wrongly, but I was getting no drive from the ecu to the coil (i.e no dwell) so one of the suspects was the pickup.
As it turned out it was the ignition switch which was causing the problem - when turned to the starter position the ignition voltage dropped out. If I'd bumped or hotwired the car it would probably have started OK!
What's more the car now runs with the old, diagnosed-as-faulty, ecu. The mechanic who reckoned it was dead said he was seeing the coil output being held solidly to ground (it's supposed to go to ground like a mechanical contact breaker) so either the old ecu has some intermittent fault or had got into some strange state, or the mechanic was mistaken. I'll hold onto the new ecu until I'm sure the old one is stable, but unless ATP will give me anything back for the new one it'll be up for grabs if anyone's interested.

It's worse when you can't get into the T-shirt anymore :-)
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sat, 5 Jun 2004 15:37:54 +0100, "John Stumbles"

Hi,
My Dad had a similar problem with a Honda Legend, got a new ECU from ATP and it eventually turned out to be a faulty crankshaft sensor.
In his case what didn't help was he was given a wrong diagnosis from the Honda main garage, who got their fault codes wrong and told him the ECU needed to be replaced!
Still they made it up by fitting the new sensor free of charge, but he was 350 out of pocket on the ECU.
The lessons learned in his case was that you need to get a second opinion from someone who is really competent on this type of thing.
A lot of mechanics probably don't have the necessary fault finding skills when diagnosing these types of faults.
cheers, Pete.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

problem -

I'd
being
give
Agreed: I got a lot more info - including what I needed to resolve the problem - from Mike at ATP. Ironically, I wouldn't have got that unless I'd thought the ecu was a goner and had to look for a new one (and had the nous to ask here instead of shelling out about double to the local factors and so been put on to ATP). I haven't spoken to the mechanic but as I reported he said the ecu coil output was short to ground, so either it was so and has cleared itself (which bodes ill for long-term reliability of the old unit) or he was wrong. If he was wrong that's a bit like misreading the fault codes rather than a misdiagnosis based on a misunderstanding of the workings of the system: when I told him the new unit definitely wasn't short to ground but still wasn't working he admitted he was pretty much stymied.
Anyhow bottom line is I'm probably about 120 down by buying the ecu if I needn't have plus about a day's faffing around (half of which is lost earnings) and although that's a pain if I'd taken it elsewhere it could have been an extra 100 for an ecu from elsewhere plus godknows how much labour if they'd had similar difficulties diagnosing & fixing it, so it could have been worse. :-|
Also seems from your dad's experience I could have suffered more on the cost of the ecu!
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sat, 5 Jun 2004 19:28:51 +0100, "John Stumbles"

I spoke to my dad and the name of your contact at ATP rings a bell. In his case a local garage was doing the work and bought it through an agent of ATP so it became a bit of a four way wrangle!
Also the new ECU didn't even work as well as the old one plus the original part number on the new one didn't tally with the previous one, which led him to think he'd been supplied the wrong unit. In fact the ECU stores the crankshaft sensor setting in memory, so it operates (quite well in fact) without the sensor in a 'limp home mode'. However a new unit won't have this setting in memory and will have a lot of trouble idling without sensor.
What helped him was he phoned another ECU place and they told him straight out that fault code was for a faulty crankshaft sensor. He then got in touch with Honda head office who confirmed this, and went back to the main dealer and got them to contact head office too, which is when they realised they'd screwed up. They claimed that the VIN number didn't match the year of the car, sounds a bit iffy though.

Yes, if ATP still do a checking service as another poster mentions that would be very useful.
cheers, Pete.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

IMHO, it's impossible for it to work with a totally US crank sensor - this tells the spark when to fire. Without any trigger from that there would be no spark - and no way of timing one so the engine could run at all. It could of course guess if only some of the information from the sensor was missing.
--
*I used up all my sick days so I called in dead

Dave Plowman snipped-for-privacy@argonet.co.uk London SW 12
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 07 Jun 2004 00:57:37 +0100, Dave Plowman

Hi,
Just looked it up on the web and there's also a top dead centre position sensor (and a No 1 piston position sensor too!):
<http://www.acura-legend.com/vbulletin/showthread.php?s=&threadid 72>
so it's possible the car can run on the other two sensors alone. The car had trouble idling with the new ECU, as apparently the ECU can use some memorised reading if the crankshaft position sensor fails.
cheers, Pete.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

It can't strictly speaking be a crank position sensor if other devices have this function. It presumably measures engine speed and the rate of change of that. Nor can any crank position sensor give an accurate indication of where a piston is on a four stroke engine - you need a cam position one for that. Although many engines simply fire two plugs at once - the so called wasted spark system - to avoid the need to know which piston is actually near the end of the compression stroke.
--
*A cubicle is just a padded cell without a door.

Dave Plowman snipped-for-privacy@argonet.co.uk London SW 12
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

The Honda V6 uses two other sensors, a Cam phase sensor, and a tacho sensor, both in a dissy-like housing, so yes, it will run with a U/S crank sensor. 99.9% of other cars won't though, as you say.
Tim..
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Yup - my BMW will run too with a faulty cam sensor. It's just that if it can, it hasn't got a crank *position* sensor - simply a crank movement sensor of some form. But if I remember earlier this thread, the magnets on the flywheel included a group to determine the position, so I'd say it's highly unlikely it would run at all with this sensor open circuit. Of course they frequently go intermittent - usually with heat.
--
*When it rains, why don't sheep shrink? *

Dave Plowman snipped-for-privacy@argonet.co.uk London SW 12
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.