Fixing a laboratory waterbath

[vrasp]

Re: Fixing a laboratory waterbath

Hey guys,

Took a 24h break from this. Now I m back at it.

I powered on the waterbath, the display lit up for a second then turned off. Moving it around did not make it turn back on. After some measurements, I found that the fuse on the power supply PCB blew. Below is the PCB. The culprit is the fuse above the "15v" text in green. I removed it to double-check, and indeed there is no continuity.




Do you know why it might have been fine yesterday and blew as soon as I turned on the machine today?

Is this an indicator that something in the circuit is pulling too much current?

Can I bypass the fuse? I guessing that's risky.

[eccerr0r]

Re: Fixing a laboratory waterbath

Never bypass fuses, they are there for a reason. It should have also helped protect that trace that blew in the first place, I still don't know what's the story behind that.

Supposedly this supply - the 15V -> bridge ->capacitor -> 7805 regulator, though I did not track it carefully, powers the display meter. Having that fuse blow will shut off the meter. Now determining what's causing the overload is another problem, must be something eating a lot of power somewhere on this rail... if it blew right away, this likely is a dead short which is slightly easier to find...

A "DBT" tester would be helpful here - but not in the regular sense. You should get a 0.5A-1A 12V (6-12W, not a headlamp bulb or brake light, but perhaps a festoon light or side marker) car lamp and use it as the "fuse," so you don't burn too many fuses while searching for the short.

[vrasp]

Re: Fixing a laboratory waterbath

Never bypass fuses, they are there for a reason. It should have also helped protect that trace that blew in the first place, I still don't know what's the story behind that.

Supposedly this supply - the 15V -> bridge ->capacitor -> 7805 regulator, though I did not track it carefully, powers the display meter. Having that fuse blow will shut off the meter. Now determining what's causing the overload is another problem, must be something eating a lot of power somewhere on this rail... if it blew right away, this likely is a dead short which is slightly easier to find...

A "DBT" tester would be helpful here - but not in the regular sense. You should get a 0.5A-1A 12V (6-12W, not a headlamp bulb or brake light, but perhaps a festoon light or side marker) car lamp and use it as the "fuse," so you don't burn too many fuses while searching for the short.

The trace was my fault (again). The board was touching the metal of the waterbath top.

[petehall347]

Re: Fixing a laboratory waterbath

should be able to trace the short with no power applied .

[redwire]

Re: Fixing a laboratory waterbath

I would check the filter capacitors, they might see higher voltage with the new transformer, or just died from old age. FRAKO are not a brand that lasts.
With the power off, put on ohmmeter across each one to find where the short-circuit is. The tantalum capacitors could also be a lesser possibility.

[vrasp]

Re: Fixing a laboratory waterbath

Petehall347 yes. I ve been checking each pin for anshort to ground. I might put a fuse back on the board and turn everything on without the display board and see if it blows again.

I would check the filter capacitors, they might see higher voltage with the new transformer, or just died from old age. FRAKO are not a brand that lasts.
With the power off, put on ohmmeter across each one to find where the short-circuit is. The tantalum capacitors could also be a lesser possibility.

Thank you redwire. I ll try that tomorrow.

[vrasp]

Re: Fixing a laboratory waterbath

Looking at the image below, we can see the 2 traces from the top 2 pins go straight down to the bottom right corner then straight to the display board on the botton of the image. This is the same circuit as the fuse. It doesnt seem to go through any capacitor. Does it?



Now there is something i d like to understand. If indeed it goes straight to the display board, is it safe to assume that the short would be on the display board? Or is that not necessarily the case?

[eccerr0r]

Re: Fixing a laboratory waterbath

disconnect it and see if it goes away?
Can't assume, need to check.

[vrasp]

Re: Fixing a laboratory waterbath

I would check the filter capacitors, they might see higher voltage with the new transformer, or just died from old age. FRAKO are not a brand that lasts.
With the power off, put on ohmmeter across each one to find where the short-circuit is. The tantalum capacitors could also be a lesser possibility.

All three capacitors on the green/blue board are working fine. I took them off the board and measured the resistance. It increased slowly as expected.

I d like to try power everything on without the display board but I need to get a fuse (or something that would serve as a fuse) before.

[petehall347]

Re: Fixing a laboratory waterbath

best check the AC voltages really .

[vrasp]

Re: Fixing a laboratory waterbath

I put a 12v 5w car light I found. I turned the waterbath on and measured 5v DC after the bridge rectifier. I also tried with the display board connected. Even with the display board connected there is still 5v on the output of the bridge rectifier. There doesn't seem to be any issues (aside from the fact that the display still doesn't turn on).



By the way, the car light does not turn on. I suppose there is not enough watts going through it to light it up. I suppose if it were broken there wouldnt be continuity. Does that mean the current drawn is very very small?

Why did the fuse blow in the first place but now it works? Could be my fault. Maybe I accidentally touched something with my multimeter and shorted it (just a hypothesis).

The display still doesn't work though.

I measured the voltage on each pin that goes from the green/blue board to the display board (without the display bosrd connected):

The pins are on the bottom of this picture. From left to right.



Pin 1: 6v
Pin 2: 3.6v
Pin 3: 11v
Pin 4: 6.14v
Pin 5: 6.14v
Pin 6: 0v
Pin 7: 5.25v
Pin 8: 0v
Pin 9: 0.06v
Pin 10: 8.6v
Pin 11: 4.2v
Pin 12: 1.4v

Anything suspect? Pins 9 and 12 maybe. I think 11 is supposed to be 5v since it s coming straight from the bridge rectifier of the PS board.

Remember, thise measurements were done with the display board disconnected (in the picture its connected).

It wasn't easy to measure. The board is dangling and I try to be super careful. I will double check those voltages tomorrow.

[eccerr0r]

Re: Fixing a laboratory waterbath

You have a very strange problem, it can go from working -> open -> shorted -> working... you'll have to monitor what conditions cause it to go to each state.

The lamp should not light if there's no short. However the lamp should allow the 5V to go through like an intact fuse.

[vrasp]

Re: Fixing a laboratory waterbath

I have double checked those voltages I listed. They are correct.

I have not been able to purposefully recreate the short.

I m not sure where to go from here. Maybe I can put some new solder on every solder point of the board.

If you have any suggestions let me know.

[eccerr0r]

Re: Fixing a laboratory waterbath

An intermittent problem - one of the toughest ones to debug. You'll have to go through each board manually, use like a stick or something and carefully poke at items until something causes a reaction.

I did this on one of the radios I had... had an FM stereo that didn't produce stereo - AM was ok. I found out if I flexed the board, the FM would work. Not having a schematic, I started poking components and found a coil that was intermittent. I reflowed the solder and boom, FM started working like new once more.

[vrasp]

Re: Fixing a laboratory waterbath

Finally was able to get something while poking around!

https://files.fm/f/6ce8up82u

From this, I would say it's either the pins from the green/blue board that are not well seated or the ones on the display board. What do you think?

[vrasp]

Re: Fixing a laboratory waterbath

I put new solder on the display board joints. That did it!

Now back to the other issues. I think the temperature was stuck to 35.7 celcius. Looking into that.

[vrasp]

Re: Fixing a laboratory waterbath

I m still getting some issues with the display. It s sometimes dim or I have to move it slightly. It seems to be caused by a misconnection between the long pins and the metal part inside the white plastic display board connector. I noticed there is some sticky matter on the display board. I cleaned it off but I cant clean it inside the holes of the display connector.

How can I clean this thoroughly and make sure those long pins are in contact with the metal inside the female connection of the display board.

Basically I want to clean the inside of these holes.

[eccerr0r]

Re: Fixing a laboratory waterbath

As an off-colored joke, yes I think those connectors are tired of being f**ked from behind. I suppose that's why not many people design boards with that kind of connector, not only the manual labor required to assemble them.

That connector between the blue and display board probably is the most stressed from the vibrations, not sure if a mere cleaning is sufficient, you will likely need to get creative to renew those contacts. The similar connections from the display board to the LED board probably don't feel much vibration and thus have persisted.

[vrasp]

Re: Fixing a laboratory waterbath

As an off-colored joke, yes I think those connectors are tired of being f**ked from behind. I suppose that's why not many people design boards with that kind of connector, not only the manual labor required to assemble them.

That connector between the blue and display board probably is the most stressed from the vibrations, not sure if a mere cleaning is sufficient, you will likely need to get creative to renew those contacts. The similar connections from the display board to the LED board probably don't feel much vibration and thus have persisted.

Actually, as I was about to try to figure out the temperature issue, the car light started lighting up and blinking. When I remove the LED board from the display board, it stops. I suppose there is something wrong with the LED board or the connections between the 2.

I thought of wiring the boards that have those pins but its time consuming and requires smaller wires than the ones I have. It also means I wont be able to remove it without desoldering everything if something else goes wrong.

I feel like I m taking one step forward two steps back haha

[petehall347]

Re: Fixing a laboratory waterbath

some people use pencil erasers to clean connector pins .