Fixing a laboratory waterbath

[vrasp]

It's meant to be on for long periods of time. Sometimes 10 hours. We have another waterbath (same model but without a display) that I use all day long at 80 °C for protein gelation. I have checked the heatsink of that one and I d say it's more or less as hot.

The heating is probably as fast as before. Again comparing with the other waterbath which takes at least an hour to get to 80 °C. I didnt thoroughly compare but 20 °C in 20 min is reasonable. It holds I would say close to 10L of the antifreeze/water mixture which is a lot to heat up.

I dont know how to tell if it's fake but I trust the supplier. They re the go-to place for electronics. I m not too worried about it. If it stops working I ll get one from another store.

[sam_sam_sam]

Is there a way to put a cooling fan on the heat sink safely

[eccerr0r]

heatsink fins look like it's on the outside so i dont see why not....
however the duty cycle, despite the device being used for hours at a time, the heater will not be powered up the whole time as it needs to be shut off by the circuitry else the temperature will keep going up. One it has reached the setpoint temperature the heater and thus triac should be turned off for a while.

[vrasp]

That's right. There is an LED light that is on when the heater is on. Once the water reaches the set temperature, the LED flashes, indicating the heater is turning on and off to keep that temperature constant. The TRIAC temperature would likely decrease once the set temp is reached.

I could put a fan on it but I dont want to put in the extra work. If the TRIAC dies from overheating I ll look into ways of reducing the temp. There is lots of surface area where the heatsink sits so I could also just replace it with a much bigger one.

This week I ll try to work on all the solder pads and cables I added so everything is nice, clean, and properly soldered so I can put everything back together. I wanted to keep the TRIAC fuses in there but they re too big (20 mm), especially with a fuse holder. In the end they aren't really necessary. It's more for peace of mind.

Thanks for the advice!

[eccerr0r]

10L (assuming 100% water, which is not right if there's glycol but it makes the math easier) ~= 10kg * 4184J/kg°C = 41.84kJ/°C , * 20°C = 836.8kJ to raise 20°C

So if assuming a 1KW heater, it would take 836.8 seconds = 13 minutes to raise 20°C .

So yes your numbers are in the ballpark at least, though assumptions
- it should be faster as glycol holds less heat than water.
- 1KW heater is 1KW heater, not counting TRIAC and wire losses. 1KW=8⅓A @ 120VAC RMS. At 8⅓A, almost 10W (1%) is lost at the TRIAC.

So it sounds like you have less than 1KW heater, more water, or significant heat or wire loss to environment...

[vrasp]

I checked in the manual. It says the capacity is 14 to 20L. It's close to being full so I d say 17L. That would mean ~24 min to raise to 20°C.

But again, I assumed it was around 20 min and I dont know the exact volume in the bath. I also dont know what mixture the technician put in there. He might have put all anti-freeze, or a 1:1 solution, or some other ratio. The lids are metal and absorb some of the heat, and other losses due to the system not being adiabatic. Lots of unknown variables here but like you said, it seems reasonable.

My new issue now is with the display. I get the correct temperature displayed when I turn on the waterbath. Then when I press the button to set the temp using the potentiometer, it shows random numbers constantly changing. It should show me the set temp and let me change it. When I release the button to go back to the current waterbath temp the values are not correct. Adjusting the potentiometer increases/decreases the number so that seems to work.

Turning the waterbath off and on goes back to giving me the correct temperature readings on the display.

I replaced the ICL a while ago. It wasn't easy installing it because it is so long and I had issues melting all the pads at the same time to insert it. As a result, most pins inserted through the holes as they should. a few pins did not go through and bent (like 3 pins I think). I put solder on those and made sure there was continuity between the pin and the pad on the other side of the board. I think it should be good but I ll double check.

I suspect the issue has to do with the long pins from the display board into the LED board. I had issues with the pins from the other boards and ended up removing them and putting wires instead. I dont think I have space to do that in this case. I m gonna try tinning the pins to increase their thickness and improve contact. I m open to any advice you guys may have.

Thank you!

[eccerr0r]

And aha! the manual says the RMS6-20 holds 14-20l of circulating fluid and the 1KW heater guess was spot on.

So 20 minutes to raise 20°C is extremely close to what it should be according to the math.

... and the manual also says that 60°C outside case, assuming the triac is one major contributor, is normal...

[vrasp]

Hey guys,

Just an update.

I decided to reassemble everything. I couldn't figure out why the display is showing the wrong temperatures. I checked all the connections and it all seems to be good. Maybe the ICL needs to be replaced again, or the op amp. I don't fully understand how this works so it would take me a lot of time to investigate this issue. I ve decided it doesn't matter. The waterbath works, the temperature can be controlled, and I ve added a thermometer in there to get the actual temperature (2nd picture, bottom right).

Yesterday, I had to open the other waterbath we have (RM20, 1st picture). It's the same brand but a different model (the one I ve been working on is a RMS20). I looked at the TRIAC and noticed it doesn't have a mica but it has a washer. I remembered that when I first opened the RMS20 I did not see a mica. I put 2 and 2 together and realized the reason why I was not getting continuity with the heatsink was because it has an insulating layer on it (the 2 waterbaths have the same heatsink). So no mica needed. The TRIAC only has a washer because the screw makes contact with the exposed conductive part of the heatsink inside it. All these TRIACs I blew were likely due to the washer or screws (maybe size or not well assembled). Anyways, I had an "aha" moment when I saw there was no mica in the RM20.

I ve added a picture of the RM20 (1st picture). As you can see there is no display on this model, just a knob with temperatures. How complicated would it be to install a knob like this?

Thank you!

[sam_sam_sam]

Here is a website link where you can find temperature controllers

https://www.ebay.com/sch/i.html?_fro...oller&_sacat=0

I hope this helps

[eccerr0r]

Again it's still possible the voltages were not assigned properly at the transformer, however at this point with all the other damage on the board and their associated bodge fixes, I'm not sure where to start looking for problems...

[vrasp]

Originally Posted by sam_sam_sam

Here is a website link where you can find temperature controllers

https://www.ebay.com/sch/i.html?_fro...oller&_sacat=0

I hope this helps

Thanks Sam!

Originally Posted by eccerr0r

Again it's still possible the voltages were not assigned properly at the transformer, however at this point with all the other damage on the board and their associated bodge fixes, I'm not sure where to start looking for problems...

Yes that makes sense. I m happy with the current way it works as it does everything it was built to do. Thank you for your help! I think this thread can finally go to sleep.

Thanks to everyone who helped.