Fixing a laboratory waterbath

[petehall347]

Re: Fixing a laboratory waterbath

To keep it liquid at 120°C, you must increase pressure to at least 200 kPa (so around 100 kPa gauge pressure).

[eccerr0r]

Re: Fixing a laboratory waterbath

btw how the heck do you get a water bath to 120°C, is it pressurized water?

[vrasp]

Re: Fixing a laboratory waterbath

It's not the temperature range of the bath that's the problem, the phase change motor's vibration is what kills electronics. I'd suspecting it's what killed your connectors in the first place, and perhaps if that avenue was checked first, there wouldn't be so many bodges around now...

That is likely why it worked for a week. I ve been doing a lot of things wrong since I opened this thing a few weeks ago. Even with all these issues I get excited about this and look forward to working on it and understanding better how these circuits work. I think I ve found myself a new hobby. And of course, as I ve said before, I m really appreciative of you guys taking some of your time to help me even after half of these issues I ve caused myself.

[eccerr0r]

Re: Fixing a laboratory waterbath

It's not the temperature range of the bath that's the problem, the phase change motor's vibration is what kills electronics. I'd suspecting it's what killed your connectors in the first place, and perhaps if that avenue was checked first, there wouldn't be so many bodges around now...

[vrasp]

Re: Fixing a laboratory waterbath

look closely at these places ...

One problem I see is that you are not getting your connection hot enough for the solder to flow and those connections are not going to be very good and you need a fuse holder ( or a fuse that has pig tail wire on the fuse ) and not solder the fuse to board like you did those are not going to be good connections

I am sorry if it sounds like I am giving you a hard time with this but it is important that your soldering joint connections are solid so you will not have issues later on with this device

When you see the strands of wire and very little solder on them this not going to be a very good connection you would be better off if used a piece of wire with the insulation still on the wire and go from point “A” to point “B” and make a good connection on point “A” to “B”

Or

You would have been better off if you had flatten the end of the wire across the the solder joint pads and solder it that way then your connection would have looked a lot better in the long run

If you can find wire wrapping wire which is one solid wire and you laid them side by side and flowed solder on that to make solder trace you would have been better off doing that

Are you using solder that has flux inside the solder if you are then the solder might not be of good quality if you are having issues with it to flow

Not all solder that has flux in it are the same in quality be aware of this back in the day when Radio Shack was every where you could find good quality solder you can still find solder on there website however I do not know if the quality is still exactly the same as it was back then

This is good advice. I d rather you tell me that and I fix it rather then having to open everything back up to fix connections in a few weeks/months. This waterbath is supposed to run (if I m able to fix it for good) at temperatures from -20°C to 120°C. The connections must be good.

I ve been under-estimating the power of flux. I will reflow solder on these connections with some flux see what I could do about that fuse.

Thank you!

[petehall347]

Re: Fixing a laboratory waterbath

look closely at these places ...

[sam_sam_sam]

Re: Fixing a laboratory waterbath

One problem I see is that you are not getting your connection hot enough for the solder to flow and those connections are not going to be very good and you need a fuse holder ( or a fuse that has pig tail wire on the fuse ) and not solder the fuse to board like you did those are not going to be good connections

I am sorry if it sounds like I am giving you a hard time with this but it is important that your soldering joint connections are solid so you will not have issues later on with this device

When you see the strands of wire and very little solder on them this not going to be a very good connection you would be better off if used a piece of wire with the insulation still on the wire and go from point “A” to point “B” and make a good connection on point “A” to “B”

Or

You would have been better off if you had flatten the end of the wire across the the solder joint pads and solder it that way then your connection would have looked a lot better in the long run

If you can find wire wrapping wire which is one solid wire and you laid them side by side and flowed solder on that to make solder trace you would have been better off doing that

Are you using solder that has flux inside the solder if you are then the solder might not be of good quality if you are having issues with it to flow

Not all solder that has flux in it are the same in quality be aware of this back in the day when Radio Shack was every where you could find good quality solder you can still find solder on there website however I do not know if the quality is still exactly the same as it was back then

[vrasp]

Re: Fixing a laboratory waterbath

the washer is a special one with a lip that goes into the triac so the screw cant touch the sides.

Yes that's what I got. It came with the TRIAC.

[stj]

Re: Fixing a laboratory waterbath

the washer is a special one with a lip that goes into the triac so the screw cant touch the sides.

[vrasp]

Re: Fixing a laboratory waterbath

Hey guys,

I ll be posting some pictures of what I am doing to make sure I don't make any mistakes.

First, the triac. I put thermal paste between the triac and mica and mica and heatsink. I put what I think is considered a thin layer. The triac is screwed to the heatsink with a washer so the screw doesn't touch the metal plate. I do not know the size of the screw but it is smaller than the triac hole. There is no continuity between the triac and heatsink or the screw and the triac metal plate. It all seems to be good but I d like to run it by you first.

I soldered the new capacitor. The way I understand this is there is no polarity in film capacitors, and also it's AC so I don't have to watch for polarity. The transformer is back where it was. I believe everything is set. I m going to get a 100w bulb and hook it up between the transformer and the PCB.

If you guys see anything abnormal please let me know. I m quite worried I either forgot something or did/hooked something wrong and do not want to have to go get new components once again.

Thank you!

[vrasp]

Re: Fixing a laboratory waterbath

Sounds good I ll do that. I might use the lamp bulb trick here because I m worried about causing a short again. Like putting a 100w bulb between the primary of the transformer and the PCB.

[eccerr0r]

Re: Fixing a laboratory waterbath

capacitor shouldn't be harmed unless the desolder job broke it.
You should use one or the other insulator, just not both.
Make sure you use that shoulder washer that came with the TRIAC. The screw must not touch the TRIAC either.

Without "the helping hands" tool, probably just mount the TRIAC and then solder to the pins. Careful not to pull on the wire after it's been soldered. Should make it a bit easier to solder. Definitely tin the wires first as you were doing.

You don't want a socket here.

[vrasp]

Re: Fixing a laboratory waterbath

Thermal paste (use VERY little just like on a CPU) should be on all surfaces that sit between a heat source and the sink where it could have touched air.... so both sides of a mica or kapton insulator.

If you have a silicone insulator, which feels rubbery, you usually don't need the paste - only the hard stuff.

In a properly designed circuit especially with TRIACs you may not actually need the paste as the dissipation shouldn't be as high as semiconductors running half on or at high frequencies.

The standard TO-220, the metal tab is connected to the center terminal of the three pins in general. If your center pin is connected to hot on your AC input... BANG dead short to ground.

However if you dirty the contact from the tab to the heatsink you might get high resistance, though you don't want to leave that to chance - use the insulator and the shoulder washer if it was insulated before.

There are all plastic coated TO-220FP that are isolated from the center pin.

Got it. Thank you!

Just got this stuff in the photo. Hopefully I m all set. I got a capacitor to replace the red one in case it broke when thebshort happened. I have to wait until next week to be in the lab.

I get an extra insulator on top of the one provided with the triac. Which one should I use?

Any tips on how to solder the triac pins to a wire? I just put flux and solder on both ends and melt it together for a few seconds then add a few layers of shrink tube (Like in the previous photo of the triac). Maybe there is a "cleaner" way to do it - like some kind of specialty connectors I can get with a female end to insert the pins.

[eccerr0r]

Re: Fixing a laboratory waterbath

Thermal paste (use VERY little just like on a CPU) should be on all surfaces that sit between a heat source and the sink where it could have touched air.... so both sides of a mica or kapton insulator.

If you have a silicone insulator, which feels rubbery, you usually don't need the paste - only the hard stuff.

In a properly designed circuit especially with TRIACs you may not actually need the paste as the dissipation shouldn't be as high as semiconductors running half on or at high frequencies.

The standard TO-220, the metal tab is connected to the center terminal of the three pins in general. If your center pin is connected to hot on your AC input... BANG dead short to ground.

However if you dirty the contact from the tab to the heatsink you might get high resistance, though you don't want to leave that to chance - use the insulator and the shoulder washer if it was insulated before.

There are all plastic coated TO-220FP that are isolated from the center pin.

[vrasp]

Re: Fixing a laboratory waterbath

I was also thinking about this and I suspect it is a pulse transformer and not a relay. Likely similar to a Schurter IL series
What is your plan for the transformer with the leads ripped out?

The ripped out leads are from the previous transformer that has been replaced since.

That trace blew because you didn't put the insulator on. You need both the kapton/mica/... insulator with heatsink compound, plus the plastic shoulder washer. If the old screw doesn't fit with the shoulder washer, you need to find a thinner screw, usually #4 or M2.5 screws are thin enough.
Since you blew that trace, yes you may have blew the (relay). You might be able to get away with an arbitrary 12V relay but I still think it's optical because relays are not fast enough.

Then again perhaps the original designer was an idiot and designed it this way... *facepalm*

BTW, the "input" side should be quite conductive if it's a relay, this would be a good test and shouldn't get damaged by the shorting TRIAC. If it were a true optoisolator it needs to be an LED input, and these would be conductive in only one direction... again if it really is a relay, this is a very poor design.

Oh and another thought... it could be a transformer. Now discerning a damaged transformer and a relay with simple tools...

Should the thermal paste be between the triac and mica or mica and heatsink? or both? Also, what is the triac metal back plate connected to inside the triac? I thought the issue was that the triac metal part was touching the heatsink and was therefore grounded, but I dont get continuity between that metal part and the heatsink.

EDIT:MT1, MT2, and the gate all have continuity with eachother (even when the triac is not on the heatsink) which I think means it's damaged.

[eccerr0r]

Re: Fixing a laboratory waterbath

Isolation is to make sure two circuits cannot interact with each other with a single additional connection. They are mainly for making two circuits that can be connected together without unexpected interactions, and the more important reason: protection from electrocuting people using the circuit.

[vrasp]

Re: Fixing a laboratory waterbath

I never thought of a transformer that way..to isolate circuit. I always just saw it as voltage step up / step down. That s so useful.

No I never heard a click. I tried to listen when I was testing it. Nothing. I believe you guys are right and it s a transformer.

Thinking further about this, if you have a short on one side which drives up the current, the voltage will go up too and have repercussions on the other side of the transformer. So the circuits are physically isolated but still reactive to one another/energically interconnected. What benefit does the physical isolation provide?

[eccerr0r]

Re: Fixing a laboratory waterbath

yep same reason for putting a relay or optoisolator there, transformers also provide isolation. Of the three, relay is the dumbest (why have a triac then?) Transformer is "ok" but trickier to use, and optoisolator would be ideal.

Now the question is whether you ever heard it "click" ... a transformer making clicking sounds is not normal.

[R_J]

Re: Fixing a laboratory waterbath

The transformer may step up the pulse from the drive circuit, but it could also be for isolation, isolating the driver board from the triac which would be connected to the ac line.

[vrasp]

Re: Fixing a laboratory waterbath

That's great! Very helpful. Do you know why they put s transformer there? What is its purpose?

Thank you!