So since fixing the radiator, a few times when i've done a short journey i've noticed steam coming from under the bonnet when i stopped. Initially i just marked it down to it being frosty, and the engine melting frost/condesnation away. Its used a small amount of coolant since fixing the rad, but i wasnt entirely sure if the topping up was simply down to small airbubbles or whatever working their way out.. However this afternoon i went to the nursery to pick up the wee man (about a mile away) and when i reversed into the bay noticed steam coming out the sides of the wheel arches and grille, and when i stopped it was whisping out the sides of the bonnet. Clearly not condesation!!

So i popped the hood, and had a look. Offside rocker cover was wet with a puddle of coolant and there was steam pouring off the exhaust manifold. No obvious leak, but the T pieces for the LPG system are directly above where the coolant seemed to be, and the T Pieces looked damp. I fired it back up, and no visible leak. Drove it to the dump, sat in a queue for 10 minutes waiting to get in, parked, drove to Greggs then home, and no more steam appeared.

So i'm presuming the T pieces are the source. They're configured somewhat peculiarly and i'd like to check before i order some new bits.

So whats there just now, is two plastic equal Tees in the heater hoses, right beside the brake modulator. They've then used some VERY stiff hose, to go from those equal tees, round the back of the engine and across towards the LPG vaporiser. About 10-12" before the vaporiser, there is a straight reducer in each pipe, and it drops down to a smaller size which goes to the fittings on the vaporiser. This smaller hose is still too large, and has been crushed a bit where it attaches to the vaporiser. I'm thinking, at a guess, the very stiff hose isnt sealing properly until it warms up and softens a bit.

What i would like to do, is buy two new metal T pieces, which go from whatever size the factory heater hoses are, down to the correct size hose that fits the vapouriser. Then run two new lengths of hose to the vaporiser.

Before i do this, i want to confirm the current odd layout hasnt been done for some reason i've not considered. I'm assuming its just been lashed up by the same moron that installed the rest of the system.

I'd also like to confirm what size the factory hose is? and also if anyone might know where i can find what size pipe my Bigas vapouriser actually needs (short of taking the pipes off and measuring)

Cheers
Kev

Internal diameter of the factory hoses is pretty much 19mm, and a lot of lpg systems use 16mm I'D.

You might be able to get brass 19-16-19 T pieces, and then run 16mm pipe to the vaporizer. I'm sure you can get 19mm equal T's though. Lpgshop.co.uk have a good range of them in brass etc. And are usually cheaper than eBay sellers.

i think the vapouriser is smaller, maybe 12mm? i'll need to measure i think!

I was looking on LPG shop, they have a 19-10-19 and a 19-16-19 but nothing in between :(

http://www.bigas.it/en/products/lpg-reducer/ri-21-double

That suggests 10mm, but i'm not sure thats exactly the same as mine, but maybe the 19-10-19 will be fine after all!

Don't use Tee's at all. You need a 19-19mm straight connector, two 19-16mm straight connectors and a good length of 16mm ID hose. In the hose from the inlet manifold fit a 19-16 connector, run a hose from that to the vaporiser. Vaporiser return hose goes to the heater inlet using the second 19-16. 19-19 goes in the heater return to connect it back as it was originally.

Doing it this way the vaporiser and heater are in series so no danger of one taking all of the flow while the other gets none.

My OMVL vaporiser uses smaller hose than 16mm too, though 16 can be convinced to (eventually) stop leaking by clamping down on the lip of the hose barb. The ID though is much smaller, and I was considering running it in series with the heater core until we found my lack of heat at idle was a blocked thermostat moulding.

Had I actually run it in series though, I'd be concerned my vaporiser would be a big restriction on flow - the surface area of the ID was comparable to the blocked stat.

yeh i dont think i want to do it in series given the tiny size of the connections on the vaporiser. I planned to measure today but its snowing so it can wait.

Thinking outside the box... The original throttle heater uses a seperate fitting on the manifold, which went to the throttle then back to the expansion tank. Thats now blanked off on my engine. I wonder if i could use that instead of T-ing the heater hoses at all?

Or maybe use that as the feed, and return to the heater return as normal.

Need to have a look at the bay.

Possibly - but from memory, the throttle body heater and it's return hose are 8mm...

Which even if you have 10mm on the vapouriser, I would be really wary about the flow through 8mm hose being enough to stop the vap from freezing up.

My Zavoli Vapouriser has 16mm hose connections, and I don't have an issue with that in parallel to the heater matrix. I get good cabin heat, and never had the vapouriser freeze up - it's always too hot to touch!

yeh you could well be right, something to investigate. Hopefully i'll get a few minutes tomorrow to have a nose around.

Hmm, all very interesting.
I wonder. If Marty is still getting good cabin heat with his heater matrix running in series with the reducer then surely there must be plenty of heat left in the coolant after it has heated the reducer?
Also, would it be more efficient to run the coolant to the heater matrix first when it is at its hottest and then to the reducer which doesn't need to be as hot?

My Reducer/Heater Matrix also run in parallel without (known!) issues - but I can definitely see where Gilbert is coming from with the theory.
The idea of running two totally separate feeds using the redundant 8mm circuit from the throttle body heater tickles me. I reckon you'd get away with an 8mm feed even though it's a big difference in flow. This is based purely on gut, without maths - so it's just BS until we work out a way to test it.

Mine's running in Parallel Morat :) but i get good cabin heat and vapouriser heat running as it is..

Parallel on my two as well. No probs with cabin heat or even icing on the single point- which I start and run on LPG.

Most LPG vaporisers use 16mm 5/8" water connections but some vaporisers (including some Bigas vaporisers) use 10mm water hose. If the reducer uses such small diameter water fittings it would of course be best to keep the reducer plumbed in parallel with the heater matrix.

It's unlikely water hose was that stiff when fitted, best advice would be to change any pipes that have gone stiff. I've seen some so brittle it was easy to snap them like a twig... This can happen if gas hose is used instead of water hose, if water hose isn't compatible with the type of coolant, or if hose has been subjected to carrying exhaust gasses (HG problems etc).

Side point - Some of the Bigas 'twin' sequential reducers (effectively 2 reducers bolted back to back using a common water circuit and gas feed but with separate gas outputs which are then usually 'Y'd together) use only a single 10mm water feed. I suppose we could work out the flow rate of LPG needed to produce 400bhp and the cooling effect of vaporising that flow rate of LPG, could compare that cooling effect to the heat stored in water and the water flow rate through the 10mm pipe.. it'd need a greater flow rate than straight calculation would suggest because the heat exchange won't be 100% efficient. I know in practice on a lot of powerful vehicles the 400bhp reducer with 10mm water feed does seem to drop in temperature more than most reducers (with 16mm feeds) when pushed to flow a lot of gas continuously.

My preference for series plumbing came from experience. I had a Classic running a single point (so can run on gas from the word go) plumbed in parallel when I got it and on a cold morning it would freeze the vaporiser within 400 yards. Changed that one to series and it never did it again. The P38 was also fitted with a single point and parallel plumbing when I got it and at idle on a cold day the heater would drop to lukewarm. Again, changing to series cured that too. It depends on how much restriction each puts in the coolant flow. If the vaporiser has more resistance, the bulk of the coolant will flow through the heater allowing the vaporiser to freeze. If the heater has more resistance, then there's plenty of flow through the vaporiser but not enough through the heater. Series equals the flow and I've never had any problems with dropping from 19mm down to 16mm and back again but I will admit that if a vaporiser has very small coolant connections it could cause too much of a restriction. In saying that though, with them plumbed in parallel I would have thought the bulk of the coolant would go through the heater with the risk of freezing the vaporiser?

It would be interesting to see what the relative temperatures of the vaporiser and heater matrix are. A Nano will tell you what the heater temperature is and an infra red thermometer will give a good idea what the vaporiser temperature is.

Sorry Marty, I got confused :)
Anyway, it seems that both plumbing designs are valid. I will say that I do notice less heat at idle, but that was a pretty extreme example (sleeping out overnight in the Range Rover and starting the engine to heat up the seats). It was also below freezing outside. That might be down to an unequal split between the two routes, or it could just be a lack of flow overall due to the speed of the water pump. The reducer didn't freeze, but it wasn't flowing much gas either.

Series or parallel plumbing might be as broad as it is long (probably a pun in that somewhere).. On the one hand if the heater circuit won't flow much volume but has a lot of pressure then series plumbing might be best, on the other hand if the heater circuit will flow a lot of volume but doesn't have much pressure then parallel plumbing might be best.

Seems in practice on most vehicles the heater circuit would flow a fair bit more water if it weren't for the restriction imposed by the components external to the engine (matrix and reducer), so in the case of series plumbing this would point to the matrix affecting flow through the reducer and the reducer affecting flow through the matrix. But on the other hand in the case of parallel plumbing the heater circuit won't flow so much more total water that the matrix / reducer won't each get less flow.

Parallel plumbing may have become the norm because there are plenty vehicles which shut off flow of water through the matrix when the heater is turned fully off or set to the full cold position (like the Honda CRV I'm currently converting which has a mechanical cable'd flow shut-off valve in the engine bay, BMW's and modern Jags that use electronic flow regulators to driver/passenger side individual matrix's, Merc's that use a pulsing solenoid to restrict total flow, some vehicles including some Mercs/BMWs have an electric circulation pump to aid flow at low rpm and can even allow the heater to work for a while when the engine is off). Any vehicle that has such flow metering valves / shut off valves has to be plumbed in parallel.

For sure if we're finding the heater doesn't work as well with one route of plumbing we can take the other route but I don't think it's clear cut. I haven't noticed a drop in heater regardless of parallel or series plumbing on a P38 but that could be because I haven't owned one! Different model 'factory fit' vehicles even from the same manufacturer (eg Vauxhall) and with the same engine / heater matrix use series or parallel depending on the specific vehicle model and year. I have 'repaired' some factory fit Vauxhalls that wouldn't run on gas because the heater matrix was clogged, obviously these were plumbed in series. Dunno what could be read into this but on vehicles that have a heater matrix in the front and another heater matrix in the back (such as my GrandVoyager) the matrix's are plumbed in parallel... But for sure my heater gets hotter just after the throttle has been blipped (rpm increased) than when the engine is sat at idle and I put this down to a flow thing (water pump speed increases with rpm).

Whether series or parallel plumbing is best might depend on all of: model of vehicle, type of reducer, condition of components (e.g. partially clogged matrix or reducer). Series plumbing does bring about the question of whether the matrix or reducer should be first in line for hot water and how hot the water will be to the second component in line under various conditions.

I think it's more a case of parallel will work on any car, irrespective of how the heater plumbing is arranged but series will only work on a limited number of cars that have a full flow heater system..For an installer doing a number of different cars it's going to be safer to stick with parallel on everything then he knows it's going to work.

All I can say is that with mine having the flow going via the vaporiser before the heater, Nanocom reports my heater temperature as around 65 degrees with a coolant temperature of just under 90, quite hot enough for a decent heater output. We normally use the analogy of water flowing through pipes when trying to explain electricity but to turn it on its head, it's just the same as resistors in parallel, whichever has the lowest resistance will flow the most current......

Yeh I wouldn't disagree with any of that, I think installers got used to fitting in parallel because parallel will work on just about anything.

Nearly used an electrical analogy myself... just that in the case of 2 resistors in series for a given flow of electric the resistor with highest resistance would be heated most and the second component in direction of flow would be unaffected by heat dissipation by the first lol ;-)

Off point a bit - On the recent BMW745 I converted I could have taken the easy route of plumbing the reducer in series or in parallel with the 10mm oil cooler / alternator water feed and I know some installers do that but I plumbed it in parallel with the 19mm heater pipes... the short pipe from the rear of the engine that goes through the bulkhead to the matrix is a bugger to access. Some vehicles have a water shut off solenoid on the engine itself (such as 4.0 V6 Cologne's in Ford Explorers, more recent Merc V8's / V12's), got to disable the shut-off on those or regardless of parallel or series the reducer will get cold when the heater is off. Disconnect the vacuum pipe on Explorers, remove the plunger or disconnect the electrical feed on the Mercs depending on type. .

Well plot thickens, maybe...

Weather has been awful, so i havent managed to actually properly measure anything up. Everythings shut anyway so no point getting soaked when i cant order any bits anyway.

Yesterday morning, the coolant was low, and the rocker cover looked slightly damp, i only had about half a litre left of coolant. I put that in, which took it about half way up the reservoir. At a guess it probably needed another half litre to get it upto the cold fill line. Then tightened the jubilee clips. They all felt pretty loose, and all nipped up nicely. Since then, no more steam, and the rocker cover is dry. Its still seems to be losing coolant though.

Yesterday i drove about 95miles all in, two ~40mile trips, two ~10mile trips and some running around local.

This morning, the coolant was barely visible at the bottom of the reservoir, just a small amount visible just at the pipe connection. I didnt want to fill it with water, so drove to asda (which was open on new years day amazingly enough) and bought a 1L "top up" and poured that in, which took it up to the cold fill line.

So from that i deduce that my 95miles yesterday has consumed about half a litre of coolant.

So wheres it going...

The weathers been pretty crap, lots of rain, but i've been really struggling to keep the windows demisted in the cabin. It was needing fan speed 5 and 24-25c to keep the front windows clear, and the rear glass stayed kinda steamy thru most of that 95miles of driving. So ofcourse that leads to the worry that its leaking internally, perhaps in the heater matrix. Oddly no particularly coolanty smell inside the car.

The carpet can soak up a huge amount of fluid - there's a couple of inches of foam underneath it. I think it's time to have a peek underneath the carpet in the driver's footwell.
Mine was dry from the top but there was nice red coolant underneath. You might see a trace of staining on the side of the transmission tunnel under the dash depending on the carpet and coolant colours.
How is your oil?