Underwurldes - Diy Auto Top Off Project

[Underwurlde]

Oh, it's only minor niggle using 100ohms - uses 'a lot' of current when the float swich is closed. I normally design battery driven devices and using 100ohms in that manner would be considered 'lazy'.

It will certainly do the job though, so just ignore my self possessed ramblings.

Now, mircowaves. Nasty bloody things, SO GET A NEW ONE.

TBH, I wouldn't want to go probing around inside one. Not only are there some nasty high energy DC voltages floating around, that energy is backed up with a lot of POWER. If it dosen't ZAP you, It WILL burn you.

I have had these kind of burns before, from large switched mode power supplies (I used to design & build many moons ago).

CHRIST THEY HURT.

It is a burn unlike no other - they are not only surface burns like scalding, these little b*****ds are point-to-point burns - the burn(s) I used to get were on my fingers started at one side through my finger, through the finger bone and out the other side. GOD thay ache for months - it is a weird feeling having a burn inside some thing.

So, hopefully that has put you off opening microwave ovens up & trying to fix them or ytou;ll bee typoting lioike theijhss for agessa

Andy

 

[sophos9]

HAHAHA.... laughing my ass off...

I really dont wanna play with the gubbins but II'm not sure if this is something simple. Everything works fine apart from the heating. There is a large diode that coming from the high-voltage capacitor which is connected to the frame.

My multimeter shows no conducting with leads either way round, the diode is an RG510 but I cannot find anything about it?

Off topic I know and apologise but life with no microwave is difficult

Cheers Andy

 

[Underwurlde]

Everything works fine apart from the heating.

That’s the bloody bit that’s worrying me! Promise me you’ll leave it alone mate! I know it’s a pain, but is it worth it? Really???

Ski’s American BOM Comments:

1/ The 1mohms POT for VAR1 is fine… I guess a trimmer is what you want? Your not wanting to easily adjust this thing with something that has an adjustable box-mounted knob then?
2/ Linear regulator is an exact mtach, so fine.
3/ Any reason why R9 & 10 are quarter watt?
4/ Good choice of relay.
5/ Transformer is absolutely fine, is panel mounted, not PCB mounted (if that’s OK?), adequately rated at 1A.
6/ Bridge rectifier will work but that’s a huge device! This will do just as well (but always check): 58K8798
7/ Advisable to use IC holders for the ICs. (They are worth it!).

Andy

BTW, the only mfrs I don’t recognize are Illinois Capacitor & Stancor!

 

[sophos9]

That's the bloody bit that's worrying me! Promise me you'll leave it alone mate! I know it's a pain, but is it worth it? Really???

Ignorance is bliss and not knowing the damage those capacitors can do means 'no brain, no pain'. Based on what you know, roger that old boy, will leave it!

 

[Underwurlde]

Andy

(BTW, change the thread sub-title now it works! )

 

[SkiFletch]

1/ The 1mohms POT for VAR1 is fine… I guess a trimmer is what you want? Your not wanting to easily adjust this thing with something that has an adjustable box-mounted knob then?

Yeah, I'm used to using trim-pot systems with our mass-specs and I'd much rather have a fire-and-forget adjustment. Do it once then hide it away from me where I'm not tempted to tweak it

2/ Linear regulator is an exact mtach, so fine.

Sweet

3/ Any reason why R9 & 10 are quarter watt?

Yeah, cause I couldn't find 2K2 in eigth watt . Didn't look terribly hard, but figured since Newark carries them, there's no harm in going with a higher watt rating

4/ Good choice of relay.

Sweet

5/ Transformer is absolutely fine, is panel mounted, not PCB mounted (if that’s OK?), adequately rated at 1A.

Yeah, I'm cool with it being panel mounted this first go-around (will be making a few of these for friends if it works)

6/ Bridge rectifier will work but that’s a huge device! This will do just as well (but always check): 58K8798

Thanks for the tip on the rectifier. When I first looked at that one, I thought it was surface mount (and I sure wasn't gonna play with that ). Further inspection proves it to be board mount, so that one looks like it'll do the job.

7/ Advisable to use IC holders for the ICs. (They are worth it!).

Got IC holders allready (prolly should have mentioned that)


Funny you didn't recognize either Illinois Cap or Stancor, they make TONS of products that we use here in the States. I also changed R7 to 100ohm 0.125W 1% as per yours/sophos' troubleshooting. Only two things left to do now... Put in the order tomorrow, and find some good float switches

 

[Underwurlde]

Float switches may be expensive.

A crappy idea that came to my mind was to stick a small magnet onto a piece of cork floated inside an open-ended tube held vertically in the tank. A reed relay taped to the outside of the tank positioned at the minimum water level would then act as a float switch....

also changed R7 to 100ohm 0.125W 1% as per yours/sophos' troubleshooting.

As also recommended, add 100nF between pins 1 & 2 of IC4 as well if I were you.

Andy

 

[SkiFletch]

k, will do Andy. Float switches I can source relatively easily, we use them often at work and I know some good places to start looking. I've had horrific results with reed switches in the past (maybe I've just used the wrong ones), so that idea's out

 

[Underwurlde]

I've had horrific results with reed switches in the past

I'm intreagued! What can possibly go wrong with a reed switch?

Andy

 

[SkiFletch]

Heh, they don't work... . We're designing an active thermal protection system for SCUBA Divers here at work. Long story short, a diver wears a tubesuit under a wetsuit, that tubesuit is perfused with hot or cold water from a backpack worn in place of a BC. The backpack contains thermo electric coolers (aka TEC, aka peltier), pumps, manifolds, and a self-contained automatic electronic control system. The system senses the temperature of the water with thermistors and ramps up or down the duty cycle of the TECs to try and reach a set temperature. One button control, power on, or power off. No software to go wrong, everything done with passives. It can either be battery powered, or surface supplied.

Problem is, how do you turn something on with a switch that has to withstand up to 300 feet of sea water? So we figured magnetic reed switches would be perfect. You could hide the switch just under the inside surface of the pack, wave a magnet and voila, pack turns on. We went through prolly 5 different reed switches and none of them would work in our application, they were all intermittent at best. In the end we found that we didn't need to protect to 300FSW since our prototype design required us to use a square shape and thus requires the pack to be internally pressurized at depth (to prevent bowing/imploding ). Therefore the power switch only has to be waterproof to a couple feet, and there's lots of those out there.

Btw, the pack runs ~30VDC @ 70A, on the diver's back underwater. Now, where's that pic of the lighting bolt again?

 

[SkiFletch]

Ok so float switches are a whopping $5 each and as such the hysteric in me was thinking about using one more as a last-line of defense to flooding the system. In case of failure in filling mode (which I realize is nearly impossible in the case), would it be ok to put a NC floatswitch just before the relay and mount that switch way at the top of the sump. That way, heaven forbid the circuits/IC fails and the sump tank is in danger of overflowing,j the switch opens and then opens up the relay, shutting off the pump. I'd realize then you'd get into that hysterisis problem, but at least you wouldn't have a tank with a crashed salinity, dying livestock, fried return pump, and a wet floor.

And one last thing. Andy, you may know the answer, but if not, perhaps sophos could run a quick test for me. What happens when the main power goes out (simulated by un-pugging it with switches on) and then comes back on? Does the unit re-start properly or will a power outage at 1am turn it back on only to drain the topoff tank?

Just thinking like a technician here looking for modes of failure and safety

 

[SkiFletch]

A bump choo, andy, any thoughts on the extra switch?

 

[sophos9]

Sorry mate, will test tonight and let you know for def. I'm pretty sure I've pulled the plug before and it just resets itself....

 

[SkiFletch]

Nice, thanks Sophos. I figured it would, but you can never be too careful

 

[Underwurlde]

Sorry for the delay....

1/ Don't see why you can't use that 2nd float switch as an emergency stop as it were.

2/ The system will come on in a reset state (pumps off). If the float switch is demanding water after resetting, the pump will start pumping for the predetermined length of time set by VAR1 and then stop as normal.

Good to always think in terms of failure modes....

Andy