Mat's BX Blog

[mat_fenwick]

I think the inlet manifold vacuum moves a diaphragm and allows more LPG into the vaporiser.

Isn't that a safety feature ? No vacuum = no gas - stops gas spewing out if the engine isn't turning over.

Quite possibly - although there is another main solenoid in the engine bay and one on each of the twin tanks which shut off without an rpm pulse.

if I can just take it back to them to re-start the self learning program again I think my problems should be over.

Nope. The system should be continually self learning, so whatever fault condition developed 'coincidentally' with the visit to the LPG specialists, must therefore still exist. Been to another specialist today, feel like giving up on the fault finding myself - my credit card limit has just been increased so I'm in the state of mind just to throw money at the problem to make it go away. Even if it comes to a new system, at least then it will (should?) be more reliable, and have better spares availability in the future. I've had some fundemental problems with the system installation pointed out to me (some I knew about like the wiring quality) so maybe that's the best option?

(TPS is shared with the petrol ECU and that is working OK - I know the signal is reaching the gas ECU as the ability to rev is regained above 3k rpm).

[mat_fenwick]

On a more positive note I've been flexible and well enough to start fitting HID headlamps to the van (a Christmas present from Lana - better than socks!)

They are abysmal despite a reasonable voltage at the bulbs, so I'm hoping that a better light source will help. Beam pattern isn't great but not much I can do about that!

[electrokid]

Nope. The system should be continually self learning,

That does make more sense.

Been to another specialist today,

Whenever you take it in they are going to have to run the engine 'til the lambda sensor gets up to temp so your workshop time costs could be OTT.

Probably ok to assume for now that MAP and TPS should be ok. If the lambda sensor is the original one with a lot of miles on it then it's probably close to being stuffed anyway so it's probably worth replacing if you haven't done so already. You'll need the right sort - Zirconium dioxide or the less common titanium dioxide - zirconium produces a voltage so can be a single wire connection using the exhaust as ground (but not having the pipe grounded properly can cause problems of course) - titanium devices change resistance rather than prducing a voltage. If you've measured the resistance with a meter and it's a zirconium that can screw up the sensor.

I think it won't 'learn' if it doesn't have a signal to learn from - which points again to the lambda sensor.

If it was me I'd 'scope the sensor output as a first to check it - I'd warm the engine (so the sensor was hot enough) then take it up to 3K and switch to gas then reduce revs very very slowly bit by bit to give it a chance to learn something (I'd be tempted to temporarily wire in a pot in the place of the TPS to get fine control).

[mat_fenwick]

If the lamda sensor was at fault the petrol ECU would illuminate its warning light (2 sensors, one shared with the gas ECU). Easiest thing to do would probably be to swap the sensor over that the gas ECU uses - new ones are 175 pounds each for genuine parts!!! At the moment though I’m in rather a negative and very grumpy frame of mind and feel like doing nothing - please feel free to ignore me! Your suggestions have been most welcome though, thanks.

Fitted a new brake light switch to the BX today for which my back is not thanking me. Worth noting that the genuine part has a brass thread, the pattern one a plastic thread which is a bugger not to crossthread when putting on the metal locknut! Still, at least the factors had one on the shelf and hardly broke the bank at 6 quid.

[electrokid]

Easiest thing to do would probably be to swap the sensor over that the gas ECU uses

Good plan for when your back allows.

At the moment though I’m in rather a negative and very grumpy frame of mind and feel like doing nothing - please feel free to ignore me! Your suggestions have been most welcome though, thanks.

No probs Mat - it is electrickery after all - which is what I'm supposed to be good at and I'm familiar with grumpy - something going on here too - put my book down to go to sleep at 2am and woke again at 5am and got up being too restless - did some bits and pieces and wrote some stuff about lambda sensors then changed my mind and went back to bed only to wake again at 11.30 ! Still got a bit of work done on the AV rack though.

Sometimes when a tooth filling is loose, toxins etc build up behind it and if released in one go can be quite ill-making. I didn't know I had a loose filling 'til a couple of hours ago when it fell out - so I'll try to have an early night and get that fixed tomorrow - such fun

[scarecrow]

I think I'd almost prefer to visit the Citroen garage than the dentist. Salt water gargles are good.

[mat_fenwick]

Salt water gargles are good.

Hmmm. I wonder whether saying that would work in other situations?

[docchevron]

Never worked for me sadly

[electrokid]

I think I'd almost prefer to visit the Citroën garage than the dentist. Salt water gargles are good.

I'm glad you put 'almost' there Steve - my local Citroen dealer is rubbish for spares whereas the one in Aldershot 10 miles away has a guy who even remembers the part numbers - they're usually NLA but he remembers them just the same. My dentist is good but doesn't work New Years Eve but it's not giving me any problems.

If you work with heavy metals or have a sudden release of them into your bloodstream such as can happen when an amalgam filling breaks loose and releases whatever was behind it (which is likely to be mercury-rich)then one of the best things to take is corriander which helps the body get rid of them. If you hate the taste of it then mix it with mint sauce. The effect will be greater if the alalgam filling was from the mid 70s to early 90s when 'high copper' amalgam was used - no polite words to describe how bad they were in terms of releasing their mercury.

Most engineering involves heavy metals to some extent.

[mat_fenwick]

Made a decision with the LPG situation - splashed out on a new front end to the system. Just wished I'd done this before we started spending money on the current system. Oh well, at least we'll have a modern system with a good reputation, I'll be fitting it myself so will do it with care and know how it works, and the UK importer is only 20 miles away (and one of their tech guys has given me his personal mobile number if I have problems fitting it).

In other news I've been working on an anti pipe freezing system. Not for me but for the horse! The thirsty bugger gets through about 5 gallons a day, and for that we're using a 1000 litre IBC feeding an outside tap (will fit a trough and ball cock at some point). Problem is the outlet pipe keeps freezing up and it's a lot of water to be carried every day!

To keep costs down I've been trying to use what I can find - for the heating wire I've used some thermocouple cable (4 Ohms per metre), a temperature controller and probe from work (0.01°C resolution is probably slight overkill) and a cheap aquarium heater to keep at least some of the IBC from freezing. I did some calculations and then erred on the safe side to find I'd need about 20V to power the heating wire. A suitable power supply was looking expensive but then I remembered we'd bought a pattern Dell laptop charger some years ago which was 19.5V. This didn't allow the laptop to boot as it didn't contain the Dell identification circuit, so was not being used. Result at a total outlay of under a tenner!

Now just to wait for some cold weather to test it…

For our house water supply I think I'll just leave a tap on for now on frosty nights - I don't fancy working on a pipe suspended over a river with a slightly dodgy back (which is much improved, as is my general mood )

[Dollywobbler]

Now just to wait for some cold weather to test it…

Um, it's going to be a balmy 12 degrees later this week!

Glad you're back is getting better. Did occur to me that bouncing around in my Land Rover probably didn't help...

[mat_fenwick]

Ah yes, I realised that once we started! But no harm done AFAIK. Plus the fact I only realised after we started meant I was already a lot better than even a week previously.

Still not sure what I did to it - I know it affected the sciatic nerve as I had a lot of leg pain - I initially suspected a slipped disc but the doctor reckoned it was more likely to be a torn muscle in my back, and the consequent inflammation was pressing on the nerve.

I was in absolute agony, so much that Lana phoned 999 as the NHS Direct advice wasn't helping - 999 call centre asked her if my breathing was normal and she answered "No" (as I was panting inbetween shouts of pain). This got translated somewhere along the line as "Having difficulty breathing" which resulted in an air ambulance call out! This was a little unnecessary, but they were very understanding and explained that certain key words triggered an urgent response - 'chest pain' and 'difficulty breathing' being a couple of them...

[electrokid]

Yes - glad you're feeling yourself again

5 gallons a day - strewth ! ! My first thought was to combine the idea of leaving a tap on with the need to supply water to Neddy - which leads me to a short (promise) story

I built the central heating controller in 2001 (now that is a long story ! ) and the header tank does not have a ballcock controlling the supply - it has a 12v solenoid which is switched via some simple electronics which senses the level in the tank from a couple of reed type level sensors in the tank. (Upper sensor switches 'water top-up' and the lower one means 'insufficient supply' and shuts the system down.)

The valve in the feed line has been adjusted so that when the solenoid is operated the flow is 1 litre per minute. There are a couple of cheepie alarm clocks on the system - one by the header tank and one in the control box in the kitchen - the batteries of these are disconnected until the water feed solenoid is powered so they only run when water is being fed to the header tank. The clock dials have been replaced with ones that say 'litres' because 1 minute = 1 litre and so both clocks display the total number of litres supplied.

This could be useful in checking the health of the system but in practice the clocks never get looked at - they were useful in setting up the system though - knowing that it contains 76 litres told me how much anti-corrosion jollop to put in.

There's a 'but' here though - the flow of just 1 litre per minute doesn't keep the pipes clean and in particular allows any muck that gets to the solenoid to lodge there and reduce the flow which screws up the calibration of course.

But you could build a timer device that operates a solenoid on full flow intermittantly - set to provide 5 gallons a day - or put a reed level sensor in Neddy's drinking trough and top up the trough several times a day via a Smiths clock timer - that should help to keep the supply clear of ice.

I would guess that you're thermocouple cable is er.. low cost but if you have any corrosion problems with it then stainless steel wire could be used but you'd need to insulate it - good source for wire is
http://www.wires.co.uk

[mat_fenwick]

Interesting ideas as usual, but the horse supply is rainwater rather than mains as it's a way from the house. And he's less likely to need his trough topped up at night, which is when there's most need to keep water flowing. With hindsight maybe I should have put in a mains supply when I dug the trench for the electricity but there you go...

I can't see any benefit to using a solenoid over a ball cock in your system though - but I'm willing to be corrected! Surely it's extra complexity just because you can?

(As an aside, when I was trying to work out heat loss from the pipe I stumbled across a forum where they were having an argument over whether a fire sprinkler system would freeze or not in certain conditions. The general consensus was that it would, and the OP was stupid for not considering putting antifreeze into the system. I was so tempted to point out that surely if the system ever needed to be used, the fire would thaw out the pipe rather nicely!)

[electrokid]

I can't see any benefit to using a solenoid over a ball cock in your system though - but I'm willing to be corrected! Surely it's extra complexity just because you can?

I think part of the problem is - if there's anything technical to be done it's almost invariably achievable with electronics - so the question is rarely 'can we do it' but more often 'which way of doing it should we chose'. I do try not to do things just because I can - I don't always get that right It obviously seemed like a good idea at the time but I honestly can't remember if there was a real reason behind it

One thing I do remember doing at the previous house - I replaced the CH drain tap - the usual thing you fit a garden hose to and open with a square key - with a 15mm gate valve - it just looked a better way of doing it.

I don't wear pyjamas so last thing one night I stripped off and went to the bathroom for a final pee - when I stepped on the bathroom carpet it was wet - the rad was leaking. So up into the loft to tie up the ballcock in the header tank (with the current system all I have to do is throw a switch) and then downstairs again to open the valve and drain the system. Back to the bathroom to dismantle the leaking pipework and reassemble with some silicone rubber (which was in the kitchen for some reason ! ). Then refill the system and fire it up to test - problem solved.

It was then I realised I'd completed the job without having to put any clothes on