coolant level sensor experiment - part 1

[rrobson]

for those of you who dont already know, i am investigating ways of monitoring the level of the coolant in the bottle. part 1 is by means of two probes that are screwed into the bottle, which act as electrical contacts, so when there is coolant present, the circuit will be complete.
this is by far the cheapest method, and is the reason i tried it first. for all the components needed (so far) not including the coolant bottles, it has cost me 5 pounds, and would be much less if i didnt have to buy certain parts in bulk.
this is showing the basic idea, and proof that the bottle can be successfully sealed well above the operating range. so heres how i got on:



this is what i received from messenger , so a big thanks to him for providing me with them. as you may be able to see from the picture, i have modified the yellow bottle slightly.



these are what seal wheels, and make for a perfect means of pressurising the bottle.



once these were fitted, i could find out what pressure the caps would hold.





unfortunately for me, both caps worked perfectly, so in order for them to hold more pressure, i put a ring of glue around the inside of the cap, to stop it allowing air past



the picture isnt very clear, but this was the result



the gauge is reading around 40 psi, which is nearly double the operating pressure of the cap and bottle. the only problem with this mod is that it still allowed a small amount of air to escape, but held it long enough to do an experiment

with the screws in place, i pressurised the bottle to a lesser amount, to see how much (if any) air escaped from around the screws. rather stupidly, i didnt leave enough space around the inside of the bottle and the screw on the first attempt, so i had to plug it with a screw and try again





i only went up to 30psi, just to see if it would hold before i filled it with water. but i found no problems at this stage.

so the next part was to do a test which were similar to normal operating conditions, so i boiled the kettle, and filled the bottle with hot water and washing up liquid, just to aid in pressurising the bottle, and show any leaks.





but again, another problem occured, this time the glue on the cap softened in the bioling water, and blew off at 40 psi. funny to watch, but meant i could only pressurise to what the cap would hold.









although the pressure was reduced to around 20 psi, the pressure and hot water proved to be no problem for the seals on the screws, which were simple red fibre washers.



just to prove there was still pressure in the tank, opening the cap allows everything to escape



and finally this is inside the bottle. the screws i used were M5 15mm long stainless steel, and i used a combination of fibre washer, steel washer, fibre washer to help seal everything. i am going to reseal the cap, and give it another go to see how it does at double the pressure, but i decided to post this just to show i havnt forgotten about it

part 2 may take some time to complete, as i am trying to make it as cheaply as possible, and is proving difficult with the system i intend to use. if you have any suggestions/ideas/improvements, please dont hesitate to say, i will answer any questions you have

[T-Cut]

Originally Posted by rrobson

The screws I used were M5 15mm long stainless steel, and I used a combination of fibre washer, steel washer, fibre washer to help seal everything. I'm going to reseal the cap and give it another go to see how it does at double the pressure, but I decided to post this just to show I haven't forgotten about it.

Interesting. I guess there are nuts screwed onto the bolts on the inside? You didn't say how they're mounted. I assume just plain holes drilled into the bottom of the tank?

The distance between the electrodes will control how well the circuit is maintained. If the OAT strength is too low, does the conductivity reduce to a point where sensitivity is poor? Have you tried the system with plain tap water? I can't recall your initial tests on this.

If it works with plain water I guess the system is reliable in that context, but there will be a maximum electrode separation which will need checking perhaps? Anyway, an interesting approach to an important issue.

TC

[rrobson]

Originally Posted by T-Cut

Interesting. I guess there are nuts screwed onto the bolts on the inside? You didn't say how they're mounted. I assume just plain holes drilled into the bottom of the tank?

The distance between the electrodes will control how well the circuit is maintained. If the OAT strength is too low, does the conductivity reduce to a point where sensitivity is poor? Have you tried the system with plain tap water? I can't recall your initial tests on this.

If it works with plain water I guess the system is reliable in that context, but there will be a maximum electrode separation which will need checking perhaps? Anyway, an interesting approach to an important issue.

TC

sorry, i have missed a few details out. in answer to you questions TC, the nut assembly is as follows:



just to add, and rather importantly, the screws bolts and washers are all stainless steel, so they wont corrode. also, the bolt is a nylock one, so it wont vibrate lose or undo. as for the holes, the screws i used are M5 (5mm) so i drilled 3.2mm holes and simply screwed them in like self tapping screws, to aid in sealing. this way, most of the pressure would be held by the screw, rather than relying solely on the washers

i tried the circuit in plain tap water first to check the conductivity using a 12v battery and 12v L.E.D and found that a successful circuit could be maintained upto 13 inch. i couldnt test any more only due to the width of the sink. oat mix seems to be as good, but can only be certain by measuring the resistance over a set distance, which i didnt do. i set the electrodes around an inch apart, knowing that this would be plenty. if you wanted to, you could put one at either side of the bottle with no problems. i have also designed a circuit which can be wired into the cig lighter



the square in the middle being a relay, NO being the normally open circuit and NC being the normally closed part. com in obviously the common supply. (if you dont know what i mean, search for relays on google)
if this is wired through the cig lighter, it will only work when the ignition is on, and the cig lighter is easy to get at, rather than trying to rout it through the ignition.

as for positioning, that is probably the only place it could go, as it is nearly impossible to get the nuts, washers onto the bolts if not directly below the filler hole.

[VVC-Geeza]

Good luck with this,it could prove an excellent safety net for us 1.8 owners inpartricular.

I will be following your progress closely

[T-Cut]

Good stuff. Does the electrode circuit carry enough current to activate a relay? I'd have thought the current flow in that part would be in milliamps. Maybe that's enough, I couldn't guess.

TC

[rrobson]

Originally Posted by VVC-Geeza

Good luck with this,it could prove an excellent safety net for us 1.8 owners inpartricular.

I will be following your progress closely

thanks, this is only the first part, and a simple method for someone on a budget and a little technical know how. not designed solely for the 1.8 owners, i could have done with this system just before mine overheated hopefully the second part will prove to be as successful, but will be much more expensive. a little more testing should show whether it will hold the pressure / temp for longer periods of time, aiming for an hour, but have a couple of bits to gather before i can do that. the only problem i can see is if the washers deteriorate during running periods, it will allow a large amount of fluid to escape, so i may have to use rubber washers, but i have confidence in the sealing method.

[rrobson]

Originally Posted by T-Cut

Good stuff. Does the electrode circuit carry enough current to activate a relay? I'd have thought the current flow in that part would be in milliamps. Maybe that's enough, I couldn't guess.

TC

no neither could i, im not an electrician by trade (everyone who reads this is now beginning to worry) but i do know a great deal about it, i have done a lot of studying about it and designed many systems, latest one being a reverse sensor system for a forklift truck.
it is easy to put a test metre on, or even try a relay. the distance between the electrodes is relatively small, so i dont see it being too much of a problem. will cross that bridge when i get there. my initial plan was to have an L.E.D on the dash, but as with the coolant temp gauge, if something stays the same for long periods of time, you tend not to notice change, so a light going out would probably be overlooked. as for a buzzer...

this and the hans gauge together would be the ultimate in hgf prevention, so im hoping it will work as i want it to, and at a minimum cost

[T-Cut]

If you get this to operate as you want, you may have to consider if there's any effect from electrolysis of the water. At 12 volts nominal, there will be a small amount of hydrogen/oxygen generation and maybe an effect on the service life of the OAT, which is ionic I believe. It may be unnoticable, but something to consider when passing an electric current through the coolant.

TC

[sikelsh]

This is great stuff, cracking reading.

Would something like this work?



http://www.virtualvillage.co.uk/wate..._medium=shcomp

Simon

[eskezz]

Good grief. I am not easily impressed these days-but for the record I'm impressed. Good luck-hope you succeed.