Mat's BX Blog

electrokid · Post by **electrokid** » Fri Dec 24, 2010 10:47 pm

One fault that could cause that problem - if the rpm sensor is a magnet and coil pickup.

With such sensors not only does the frequency increase with speed but voltage increases too. The electronics invariably has a voltage input threshold below which a signal is not recognised.

If such a sensor becomes loose or becomes too far away from the magnet or if the magnet has lost some of its strength then the voltage output (therefore particularly at slower speeds) can become too low for the electronics to 'see' it.

I won't go into details here - but if the magnet becomes wet it can lose up to 20% of its strength every 24 hours.

Have a good one

Post by **mat_fenwick** » Mon Dec 27, 2010 1:28 pm

Interesting theory Brian, but the LPG system simply relies on the ignition pulse taken from the LT side of the coil (and is ruuning perfectly on petrol). The only difference was that it was running acceptably on gas, it was taken in for fine tuning of the existing LPG map and now is undriveable on LPG. And I had to pay for the privilege, not only for the 'work' done, but also by having to run on petrol for the foreseable future.

It will cost around 80 pounds (fuel/labour) to just get back to the place before they attempt to fit any new parts to it.

Well pissed off and not in the Christmas sprit very much. Still no water either.

Post by **mat_fenwick** » Mon Dec 27, 2010 9:41 pm

Woo hoo! We now have water!

Funny how something like that can change your outlook - feeling happier all round. Will try and whinge less in future...

Dollywobbler · Post by **Dollywobbler** » Mon Dec 27, 2010 10:14 pm

We have lots of water! Still falling out of the sky. Still lots of snow on the ground. Interesting conditions under-foot! Quite excited at the prospect of NOT de-icing the cars tomorrow...

Glad your water's back on - I could smell you from here!

scarecrow · Post by **scarecrow** » Tue Dec 28, 2010 9:25 am

Borehole in the summer?

electrokid · Post by **electrokid** » Tue Dec 28, 2010 12:05 pm

Glad to hear your water is back on - we do like to see a happy Mat

Being a bit anal about winter 'survival' I did monitor the temperature of the incoming water supply during the long freeze last year and it went down to 0.5°C so we were somewhat close to having supply problems then. But then 'latent heat of freezing' (is that the correct term ?) should have meant the pipes staying clear - the amount of heat needed to turn ice at 0°C to water at 0°C - the same amount of heat will raise the same amount of water from 0°C to nearly 80°C !

but the LPG system simply relies on the ignition pulse taken from the LT side of the coil

Yes I realised it probably did that halfway through writing the post - but carried on anyway just in case

Getting ripped off is damned annoying particularly where the kit involved isn't widely supported - have you done a web search for the make of LPG system you have fitted ? There might be a user group or someone who has found a way to access the mapping. Do you know what sort of port is fitted to the LPG ECU ?

scarecrow · Post by **scarecrow** » Tue Dec 28, 2010 6:37 pm

electrokid wrote:Glad to hear your water is back on - we do like to see a happy Mat

Being a bit anal about winter 'survival' I did monitor the temperature of the incoming water supply during the long freeze last year and it went down to 0.5°C so we were somewhat close to having supply problems then. But then 'latent heat of freezing' (is that the correct term ?) should have meant the pipes staying clear - the amount of heat needed to turn ice at 0°C to water at 0°C - the same amount of heat will raise the same amount of water from 0°C to nearly 80°C !

Pipes will only freeze near to, or above ground. Go a few metres down and it stays pretty constant at about 10 Deg C all year round... If I add the proviso, "I think" then I won't have to go and look it up

Post by **mat_fenwick** » Tue Dec 28, 2010 8:27 pm

I've found an LPG forum which has been of some help - but it was a fairly short lived 'stop gap' system (BRC Just Heavy) while the development of gas systems was in a pretty quick phase so info is pretty sparse. The comms port uses standard TTL levels and I’ve made an RS232 to TTL converter (but that circuit has developed a fault which I’ve not fixed yet - can’t even echo characters using Hyperterminal!) But software is the other issue, I’ve only been able to get hold of the software for the later generation of the system which of course is incompatible...

But while I’ve been recuperating (plenty of spare time!) I’ve given the whole thing some thought and may have a possible cause for the problem - I‘ll describe it to see whether it makes sense to others. After getting the Disco back home I spotted a split vacuum hose from the inlet manifold to the vaporiser. I think the inlet manifold vacuum moves a diaphragm and allows more LPG into the vaporiser. The hose I know wasn’t split to start with as I reconnected it when I put the rebuilt vaporiser back in - I remember it being a tight fit of the stub it pushed onto which is where it had split. The gas ECU is capable of self learning over (I believe) a preset time period; after that it will continue to run with the parameters it has learnt. With me so far?

Now what the LPG specialist was telling me is that there was insufficient gas pressure for some unknown reason and hence the system would not run properly. I’m not thinking that if the vacuum wasn’t moving the diaphragm sufficiently there indeed wouldn’t be enough gas pressure and during the self learn period the ECU would be telling the stepper motors to open fully at anything above idle to allow more gas through (to no avail as there would be insufficient pressure present). Now the self learn is over and the vacuum leak fixed I believe the steppers will be too widely open at anything above idle, and there will be the gas pressure present to supply more gas than is actually needed. This is confirmed (to my mind at least) by the fact that easing off the throttle actually raises the revs slightly (as the amount of gas admitted is better matched to the speed of the engine). Also by the fact that it smells very rich, has used a lot of gas just in the playing around period, and if you can get the engine speed above 3k rpm it manages to rev well. Does all this sound plausible? (If I am managing to make sense that is!)

Ss, if I can just take it back to them to re-start the self learning program again I think my problems should be over. (Although it almost seems too simple a problem!) If it is that I’ll me more than a little annoyed, as the split was plain to see and I would hope/expect a specialist to check and/or work out likely causes for problems themselves.

Dollywobbler · Post by **Dollywobbler** » Tue Dec 28, 2010 9:05 pm

Reading that made my brain hurt, and I think I've re-evaluated whether I want LPG on my 90! (15mpg on the first tank of unleaded, not quite in BX diesel territory...)

In other news, the light bar is removed and ready for collection/delivery...

Post by **mat_fenwick** » Tue Dec 28, 2010 9:41 pm

Dollywobbler wrote:Reading that made my brain hurt, and I think I've re-evaluated whether I want LPG on my 90!

Is your 90 on carbs or injection? This affects the type (and complexity!) of LPG kit that can/should be fitted. (From memory I didn't think they made a factory 90 V8 although I could well be wrong!)

Dollywobbler wrote:In other news, the light bar is removed and ready for collection/delivery...

Ooooh, swap you for some firewood?

Dollywobbler · Post by **Dollywobbler** » Tue Dec 28, 2010 9:46 pm

It's twin SUs. They did make factory 90 V8s and apparently, they're worth a premium! I didn't pay a premium...

Firewood sounds good. Should have got a 110...

Post by **mat_fenwick** » Tue Dec 28, 2010 10:02 pm

Dollywobbler wrote:It's twin SUs.

Then a simpler (i.e. less to go wrong...) system would suit you fine. IF you’d be doing the mileage to justify it.

Ironically the system I fitted to the old Disco is probably a little too complex for a carbed V8 (unless you fitted a lambda sensor) but (with the benefit of hindsight!) not ideal for an injected V8.

Post by **mat_fenwick** » Tue Dec 28, 2010 11:00 pm

Re the water situation - I think a borehole is out on the grounds of cost. We are actually on mains water so not that much in the wilds! If we’d left a tap on (like we have the previous few winters) I’m sure we’d have been OK, as the water only freezes (we think) over the section where it is above ground and attached to the bridge leading to our place. Moving water would not have the chance to drop below freezing during the 15 metres or so it is crossing the bridge. But being human we forgot to leave a tap on, and I’m sure will do again!
So the options are as follows I think:
More lagging on the exposed section of pipe (kind of ruled out as the cold will get in eventually, and with the lagging it has at present will freeze up within a couple of hours in the temperatures we’ve been having. Plus if it does freeze will take longer to thaw out...)
Or more likely, electrical heating wire, thermostatically controlled (surprisingly cheap!)

But the latter option does have problems - the bridge is a good distance from ours and not on our property, so we’d need a lot of cable (armoured = expensive) plus the forestry want an extortionate amount of money for us to put an underground cable in. Fortunately when we installed the phone line BT paid them for the rights, all we did was hire a digger and dig the trench.

So we could:
a) Just run a temporary extension lead to the bridge during the winter
b) Put our generator down there to power it when it’s cold (expensive in fuel plus there’s the likelihood we’ll forget to do it or can’t be bothered)
c) Put in heating cable of a rating far in excess of what’s needed simply to keep the pipe from freezing (roughly only 5 watts/metre in an ambient of -15 degrees C) and use it to actually thaw out the pipe in the event we do forget to leave a tap running overnight.

Philip Chidlow · Post by **Philip Chidlow** » Wed Dec 29, 2010 12:12 am

I am probably not speaking sense, but would it be possible to thread something into the water pipe that could be heated as required?

Depends on the bore of the pipe though.

electrokid · Post by **electrokid** » Wed Dec 29, 2010 1:09 am

If I add the proviso, "I think" then I won't have to go and look it up

and neither will I Steve

Actually my local mains is around 3 foot underground and when measuring the temp I left the tap running long enough to be measuring the temp of the local main - so I was close to leaving-the-tap-running territory.

and I’ve made an RS232 to TTL converter (but that circuit has developed a fault

Presuming comms port on the computer is RS232 and the ECU is 5v comms - on the transmit line from the computer you can usually get away with a series resistor - say 1.5K - and a couple of diodes on the '5v' side to stop the signal going further than <0, >5 volts. On the computer's receive line you can usually get away with a direct connection because relatively modern computers will usually 'see' a 5v level ok - in the past I have set up hyperterminal and connected a test probe to the receive pin so I could 'go looking' for comms. But obviously a dedicated converter will be the better option.

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.

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

Not a bad plan if it doesn't cost an arm and a leg - could be worth checking the lambda sensor output beforehand on both petrol and gas even if you have to do both tests at 3000 rpm to compare them under the same conditions - 1.1 volts for correct mixture rings a bell but I'm not sure if thats the average of a square wave or just the 'high' signal - the details for your particular sensor will be on the web. Worth checking the manifold pressure sensor too - and check that the throttle pot signal is getting to the gas ECU ok - if the throttle pot is only there for the gas ECU (and petrol carbs are driven by a normal cable) the pot may be knackered - but you've probably thought of that.

So we could:
a) Just run a temporary extension lead to the bridge during the winter
b) Put our generator down there to power it when it’s cold (expensive in fuel plus there’s the likelihood we’ll forget to do it or can’t be bothered)
c) Put in heating cable of a rating far in excess of what’s needed simply to keep the pipe from freezing (roughly only 5 watts/metre in an ambient of -15 degrees C) and use it to actually thaw out the pipe in the event we do forget to leave a tap running overnight.

d) Arrange a temp sensor to open a solenoid valve (at minimal flow to drain) when the temp falls below a certain level for a certain time.