Coolant pressure sensor

[Driver+Passengers]

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.

Agreed. I'll do load tests with this stat. Repeat tests with genuine stat, before I go for gasket, etc...

[Driver+Passengers]

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

Not quite...



I let cold start do it's thing, and then let it run for another 10 minutes. The 3mins at 2k, 2 minutes idle. On the fourth round of revving, fans went on. Left it for five minutes, another 3 minutes of revving, fans on. Left it for five minutes, 3 minutes of revving, fans on. Once fans went off, put heaters and aircon on for a couple of minutes and then switched off.

Max pressure around 17psi (?). Road test next.

This is looking good to me. I'll go and wire up a 5V regulator and see if I can get it mobile.

[Northern Bongolow]

can you put the pressure readout back to bar and pounds mate. or add it to it.

[Driver+Passengers]

can you put the pressure readout back to bar and pounds mate. or add it to it.

Note: secondary y-axis on the right.


I will be doing a road test later today.

If I don't have a head/gasket leak, then it must be down to initial fill level. Oh, for a level sensor!

[Northern Bongolow]

cheers matt, and good luck, looking at the pressure figures take your bleed stuff on the road test. .

just thinking mechanically the temps look ok its the pressure that is continuing to rise, you think this points to the head, but, i think its the stat.

(as we have been arguing for weeks) lol.

give a very knowledgeable chap on here a shout, (mel) the great pretender, he has some interesting thoughts on the pressure of the coolant system on a bongo. he is very well versed on pressure, speed, pipe sizes and flow rates of liquids.
with this in mind i see your problem as a failure due to the stat not altering the internal flow of the engine/coolant from the smaller diameter pipes (high flow rate, high pressure) to the (low speed, lower pressure) pipes. pressure inside the system is needed though to lower the boiling point, but its about controlling this.
maybe mel would pm some info.



ps. hope im not talking crap again.

[Driver+Passengers]

cheers matt, and good luck, looking at the pressure figures take your bleed stuff on the road test. .

Ooooh, yes!!!

I'm not ruling out the stat, yet, but the second test (idle) clearly showed a step change in the rate of temperature rise once the stat opened (started to open) and the radiator was brought into circuit. A knee in the graph between a steep slope and a shallower slope, if you like. It's certainly not the stat failing to open, though precisely what it's doing I'll never really know without x-ray eyes.

Not sure whether to get the level on low before this road test, or leave it higher to try and increase the likelihood of recording a blow-off...

I'll drop Mel a note after the road test, when hopefully I will have answered some of my own questions, or more likely than not raised further ones!

[Rhinoman]

I'm pretty sure that the temperature bouncing up and down is the result of the thermostat cycling, its opening at 90C and closing at 80C which sounds about right. The pressure chart is following the thermostat and peaking at 17psi - isn't that the rating of the cap?

[Driver+Passengers]

I'm pretty sure that the temperature bouncing up and down is the result of the thermostat cycling, its opening at 90C and closing at 80C which sounds about right. The pressure chart is following the thermostat and peaking at 17psi - isn't that the rating of the cap?

The fans come on at 91-ish and reliably drops the temperature by 10 degrees before the fans go off again. So the bounce in me increasing temperature with revs and the rad fans bringing it down again. The stat begins to open at 82-ish, continues to open until the rad fans come on, and by the time the rad fans switch off the stat should be fully closed.

Re pressure values: the offset voltage for atmospheric + head of water was determined three days ago, so I may be a little off by now. The cap is 1.1 bar which is 15.95psi. So give or take a psi or two, I was up at pretty much cap rating, though no blow-off was heard and no moisture was present on the tissue under the overflow. Season with a pinch of salt, to taste...

The change in rate of temperature rise during the second (no load) test seemed pretty conclusive - stat opens at somewhere around 82psi and introduces the radiator into the system - temp continued to rise at a consistently slower rate.

I've been out in the van this evening and I'll post the latest graph, shortly. I expect it to look reassuring.

[Driver+Passengers]

Finally, a road test...



0-10 minutes, cold-start.
10-20 minutes, 6 miles @40-60mph
20-33 minutes, 16 miles out @70mph
33-46 minutes, 16 miles back @70mph
9 miles detour @50-20mph, taking in several roundabouts and a hill
3 minutes stationary.


Most worryingly, on our return I let it idle for a minute or two, before putting heaters on full and switching A/C on. A/C did not raise the revs, nor turn the rad fans on. To investigate further.

I'm getting less convinced that I have a problem. As I've said before, I would like to install a second pressure sensor in the header tank. I'd also like to increase the sample rate from every minute to perhaps every 10 seconds. The temp sensor I'm currently using only samples every ~45sec anyway, and you can see the lag in temp on these graphs. I'd also like to take it on a longer run - 2 or 3 hours on the motorway.

[Northern Bongolow]

[peterrc]

I'm getting less convinced that I have a problem. As I've said before, I would like to install a second pressure sensor in the header tank. I'd also like to increase the sample rate from every minute to perhaps every 10 seconds. The temp sensor I'm currently using only samples every ~45sec anyway, and you can see the lag in temp on these graphs. I'd also like to take it on a longer run - 2 or 3 hours on the motorway.

Hi Matt, I have a spare expansion tank that you can have, and replace when convenient, if you want one quickly.
Peter

[Driver+Passengers]

I'm getting less convinced that I have a problem. As I've said before, I would like to install a second pressure sensor in the header tank. I'd also like to increase the sample rate from every minute to perhaps every 10 seconds. The temp sensor I'm currently using only samples every ~45sec anyway, and you can see the lag in temp on these graphs. I'd also like to take it on a longer run - 2 or 3 hours on the motorway.

Hi Matt, I have a spare expansion tank that you can have, and replace when convenient, if you want one quickly.
Peter

Email sent.

[Driver+Passengers]

If you're right, all I can think is that it kept venting (and dribbled on the pavement outside your house) because I kept overfilling it.

Last night's test started with 24mm in the tank, measured with a stick. I might still be running a pound or two too high. I don't know, but at this stage it's simply a case of "monitor the situation", rather than whipping the head off which I've been so longing to do!

I have to say, I'm still a little confused, but muddling through.

[Driver+Passengers]

Graph of an event. Ady - not sure - I might have replaced the coolant header tank cap with the old one, not the one you gave me. One had a larger diameter rubber seal and the other, if I remember correctly, had a shiny intake disk rather than an orangey one. I forget, now.



Many new variables in the system now. Firstly, I'm logging to an arduino, using the arduino power supply to provide 5V to the pressure sensor. For a start, this is actually measuring 4.65V.

The upwards kick in the pressure rise at 378-ish corresponds with when we got onto the motorway. I was dropping to 50ish and accelerating back up to 70, repeatedly, staying in lockup.

The red trace corresponds with the voltage at the front coolant sensor - I haven't written the conversion routine to degrees yet, but suffice to say I got up to temp and stayed there, 80-85 degrees.

I may have got my calculations well wrong, and the max pressure indicated here of 14psi, might actually be the 1.1bar. But on previous runs, I was sitting quite happily in and around this pressure.

[Driver+Passengers]

I didn't have the sensor attached this weekend, which is a shame.

Drove up to Glencoe. First vent from the cap was just a few miles out from home after a cold start and a couple of stop offs. It then continued to vent something like every fifteen minutes, often just when dropping a gear or pulling a bit harder.

Then on the way home, heard one (1) vent in the first hour and then glanced down some while later to find I was overheating. Checked heaters they were cold, so pulled over. Digital LCD gauge measuring top stub was reading 98 degC. Both passenger and I had been checking gauges regularly, and I'm pretty sure I looked at it before the previous junction which was less than a minute previously. Not a lot of air out of the funnel, other than the first big belch out of the header tank, it bled up well.

No problems for the remaining hour or two, and no vents.

I wish I had the graphs for this weekend.