Bongo Fury

Driver+Passengers wrote:
At 36m, why does pressure stabilise but temperature continue to increase (I thought I'd hit cap pressure)?

I think you need to look at the rate of change of the ratio of the two parameters. I suspect that at 36 mins the coolant is up to running temp and has pretty much finished expanding. P/T is only constant when V is constant.

Plot P/T (remember to convert the T from deg C to K) and I think you'll find that this ratio approaches a constant with a marked flattening of the curve between 30 mins and 40 mins. I had a go at this, but your plot will be much more accurate as you have the raw data.

Oh and I forgot to say, it looks good with no noticeable lag. Nice work. Any nearby Bongos you could get a control plot from?

Also rubber pipes might be expanding a little by that temperature thus increasing the volume / lowering the pressure.

Starting with a small +ve pressure in the system (0.6psi), I ran the van at idle until the rad fans came on. Here is the graph, though note I have displayed the pressure in kPa to improve the scale relative to temperature. The highest recorded pressure relative to yesterday's atmospheric reading was 15.5psi.

tallbongo wrote:Oh and I forgot to say, it looks good with no noticeable lag. Nice work. Any nearby Bongos you could get a control plot from?

Options... a new one in the next village - I might be able to persuade them if I blank their EGR for them!

I suspect the drop in temperature at 50 minutes (when the stat opens presumably) is a combination of a bolus of cold-ish coolant from the bottom hose and the fact that the radiator was cold and removes heat from the coolant until it gets up to temperature.

I can't explain the gradual drop in pressure from 30 minutes onwards, perhaps hoses expanding. Thereafter, I'm assuming that as the stat opens further that there is an effect of introducing more flowing coolant into the system. Possibly also hoses expanding, though this would have to be due to temperature, not pressure.

This was a no-load test. Other that the cold start and fast-idle when the rad fans kick in, it was sitting at 800rpm, or thereabouts. The +ve pressure in the system at rest I think was to do with the fact that I had performed a minute at 2k RPM yesterday - the hypothesis being that I had introduced additional exhaust gas into the system at that point and it had not vented. During this evenings test, it may well have been the case that I wasn't running hard enough to open up whatever cracks or fissures there may be in the head/gasket.

Next two runs will be:
- let fast start bring up to 50-odd degrees, then raise revs to 1k5 or 2k, either continuously or periodically, and observe.
- a road test along the dual carriageway.

I should perform a sniff test following those two tests to confirm my suspicions, though it remains to be seen how conclusive those results are.

Edit: that's also the last time I record the results on notepaper and manually enter them into Excel!

tallbongo wrote:Could you mount this in the expansion tank clear of the coolant?

Having used this super-putty stuff, I'm seriously considering a bodge job to get another sensor measuring air pressure in the header tank. Towards the back where the tank narrows, probably directly on top - drill a hole to accommodate the sensor port, roll a wee sausage around the base of the port and squidge down into the hole. Then build up round the sides and feather out on top of the tank. A simple zip tie around the whole tank would provide additional mechanical strength, but unless the putty fractures of the putty/tank joint fails in any way, I can't see this being necessary. The putty is good for underwater use, up to 120degC and claims a 50kg holding strength, though for what area/thickness, I don't know.

All the talk of exercising care when intruding on the coolant system, introducing more pipe joints or compromising the header tank... I'm getting a bit braver, or perhaps gung-ho. If a pressure sensor or its mounting fails, and the circuitry monitoring it is sound enough, then what could possibly go wrong???

tallbongo wrote:I think you need to look at the rate of change of the ratio of the two parameters. I suspect that at 36 mins the coolant is up to running temp and has pretty much finished expanding. P/T is only constant when V is constant.

Plot P/T (remember to convert the T from deg C to K) and I think you'll find that this ratio approaches a constant with a marked flattening of the curve between 30 mins and 40 mins. I had a go at this, but your plot will be much more accurate as you have the raw data.

Duly done, for both test runs. Note: I've adjusted scale of P/T to P/3T, and pressure is in kPa.

Driver+Passengers wrote:
I can't explain the gradual drop in pressure from 30 minutes onwards, perhaps hoses expanding. Thereafter, I'm assuming that as the stat opens further that there is an effect of introducing more flowing coolant into the system. Possibly also hoses expanding, though this would have to be due to temperature, not pressure.

is the drop in pressure due to system flow path changing matt, relative to where your measuring. as you know the stat isnt just open/ closed it operates the bypass hose closure part (top port in the stat housing) which alters the flow of the circuits. also the sizes of the relative pipes involved affects the pressure in the system, or parts of it.
when you speed up flow you reduce the pressure, and vice versa.
you really must study all this to understand your figures.

ps your head isnt cracked, imo.

Driver+Passengers wrote:

That looks pretty much as I'd expect. Volume expansion of the coolant up until around 35 mins then a flattening off until the stat opens just before 50 mins (changing the total coolant volume in circulation).

The only thing that I don't understand is when you used the heaters/aircon at 65 mins. How good is the 5V supply, as your pressure meter output varies with Vs (Vout = Vs x {0.002421 x P – 0.00842})? I'd assume that there would be voltage changes (on the 12V side) when you start using the heating system.

Can't see why the coolant volume changes when the Stat opens, its only opening the flow at one end, the volume will stay the same, but yes the flow will be different, the initial drop in temp. makes sense as the already cool/cold coolant flows into the now opened circuit.

tallbongo wrote: How good is the 5V supply,

It's a mains wall-wart.

That rules that out then

Northern Bongolow wrote:ps your head isnt cracked, imo.

If that second test was "no load" or "low load", is it possible that I simply wasn't charging the water jacket with gas? If I'm not raising combustion pressure sufficiently to open up any crack/fissure then the system should behave like a healthy system, imo. I'm going to defer judgement until I've revved it, and taken it out on a road test. What do you reckon?

do the same test/s with a gen mazda stat in it. it should ,but may not solve your problem, but the results will reveal lots imo.