Ideas for more realistic bed temperature control

vwegert

Hello,

I have the following setup in my Delta printer (top-to-bottom):

FilaPrint print surface
spring steel sheet
magnetic sheet
6 mm aluminium bed
mains voltage heater with integrated thermistor
mineral wool for thermal insulation

As you can see, the thermistor is relatively far away from the actual print surface. With tuned PID settings, it takes more than 10 minutes for the print surface to reach the target temperature, even though the thermistor reaches the target value a lot faster. The entire setup apparently has a considerable thermal mass, and I suspect that the mineral wool contributes to the thermistor showing a much higher reading than the build surface as well.

I believe there might be two approaches to solving this issue. I could try to place a thermistor closer to the print surface - however, the only way that comes to mind would be to cut a "channel" into the magnetic sheet and place the thermistor in there. It would have to be a very small thermistor with very thin leads - 1 mm at most - and the leads would need an insulation that can cope with the print bed temperatures since it might come into contact with the aluminium plate. Does anyone have experience with a setup like this? Any other idas?

(A contact-less IR sensor would have been cool, but how it might be mounted to avoid sensing the printed object instead of the bed, I have no idea...).

The other approach might be to "hack" the control algorithm - perhaps it might be as simple as changing some parameters, I don't know yet. I'm not really confident though - this might create more problems than it solves. Any thoughts on this?

Thanks
Volker

Nemesis

I have a bed consisting of a Thermal blanket, MK2 heatbed, 4mm alu plate, brosilicate glass plate 3mm. I drilled a hole in the alu plate and have the thermistor touching the glass plate.

This works very well!

I am thinking about filling the hole up with high temp silicon gel when the thermistor is there.

In your case I would drill a 5mm deep hole in the alu put a spare thermistor there and fill it with heat resistant silicon. No idea if that would work of course.

vwegert

@nemesis said in Ideas for more realistic bed temperature control:

In your case I would drill a 5mm deep hole in the alu put a spare thermistor there and fill it with heat resistant silicon. No idea if that would work of course.

Forgot to mention that the heater is a glued-on silicone mat - I'm not sure whether I could easily remove it from the aluminium plate, and I'd rather not drill into a mains-connected heater, neither on purpose nor accidentally. The end of the year is night, but these are not the kind of fireworks I'm looking for...

richardmckenna

@vwegert I have a very similar setup on my new delta build. Stuck at work on the last day waiting for the BuildTak FlexPlate system to be delivered. The last part of my build.

I did some initial tests with mine and found that because of the large thermal mass, heating the bed and then giving it time to equalise actually gave nice consistent temperatures. Yes it means having the bed on for a while before starting the print but having the large thermal mass means it takes less effort to keep at temperature once achieved.

Another option would be to cut a small channel on the top of your Aluminium plate running to the centre. Run your thermistor along that then stick the magnetic sheet on top of that.

Richard

vwegert

@richardmckenna said in Ideas for more realistic bed temperature control:

Another option would be to cut a small channel on the top of your Aluminium plate running to the centre. Run your thermistor along that then stick the magnetic sheet on top of that.

I don't have the tools to do that properly, I'm afraid... I'll try the "moar patient moar better" approach, but it's hard

richardmckenna

@vwegert ha ha I hate waiting too. Waiting for this parcel is driving me crazy.

I'll do some more tests when I have it installed and let you know how it goes.

Nemesis

@vwegert said in Ideas for more realistic bed temperature control:

Forgot to mention that the heater is a glued-on silicone mat - I'm not sure whether I could easily remove it from the aluminium plate, and I'd rather not drill into a mains-connected heater, neither on purpose nor accidentally. The end of the year is night, but these are not the kind of fireworks I'm looking for...

Ah no. Please do not drill through the heater

With a 6mm alu bed you could in theory purchase a long 4mm drill (https://uk.rs-online.com/web/p/twist-drill-bits/5226651/), drill from the side and stop in the middle. However do not attempt this without the proper tools and patience!

dc42

With 6mm aluminium, the temperature should be fairly uniform across the bed except at the edges. So no need to drill as far as the middle.

Nemesis

in that case a normal 4 mm drill would do. Just make sure you drill it straight! You do want to drill into your heater by accident.

vwegert

@dc42 Now I wish I had a FLIR to see how uniform the temperature really is on the surface and how far in I would have to go to get a realistic measurement... A shame that these things are so expensive...

vwegert

@nemesis said in Ideas for more realistic bed temperature control:

You do want to drill into your heater by accident.

No I don't.

Nemesis

@vwegert said in Ideas for more realistic bed temperature control:

@dc42 Now I wish I had a FLIR to see how uniform the temperature really is on the surface and how far in I would have to go to get a realistic measurement... A shame that these things are so expensive...

I looked into buying one of those and from I have been able to deduce the precision of a FLIR is not that good. A spot measuring tool is much cheaper and far more precise. Do that's what I use now. It takes a lot more time to measure the whole bed though...