Changing to mains-powered heatbeds

I'm getting bored with waiting for the heatbed on each of my printers to get up to 100C for printing ABS (20+ minutes) and getting annoyed when the slight draft from the hot-end cooling fan knocks the temperature down as soon as a print starts. Because of this I'm thinking about changing both printers over to 240v heatbeds, but I don't want to cause too much disruption to them in the process. Note: I don't want to wind up the voltage on the existing power supplies - that's cheating!

Both machines have a Mk3 alumin(i)um heatbed - 12v on one, 24v on the other - with glass on top. I was wondering whether I could just stick a 200mm x 200mm self-adhesive silicone-encased 240v heat-pad to the underside of the existing bed? Disconnecting the existing 12v/24v power supply, of course! I could just squeeze them in using the existing setup, with minor adjustments.

So, my questions are:

1. Would this work?
2. I know I would need an A/C SSR - do they generate any heat while working?
3. Currently I have thermal insulation under the low-voltage heat-beds - would this be necessary with a 240v heater?
4. Any other gotchas I need to know about?

Back info: I have a LOT of experience in handling 240v equipment, including a spell where I had to assess professionally whether other people had designed adequate protection in their equipment, so I am not scared by dealing with this voltage, and with providing adequate protection, etc. (e.g. the ali plate will need a protective earth connection). On my CoreXY machine the SSR will be inside the machine's casing, but I will have to build an enclosure for my Prusa (hence the question about heat generation).

Any and all guidance will be appreciated.

Cheers,
David

1. Yes, i see no reason why it shouldn't
2. Depending on exact type yes, but usually not much. Mine don't need any cooling, they are usually rated for much higher loads than what you use them with
3. Yes, even more so if you increase the wattage compared to the old one
4. Make sure the silicone heater is of adequate size, if it is to small it will heat unevenly. Also don't go overboard with power, overshooting and overheating are not much better than lomng waiting times.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

If the existing heater is stuck to the aluminum using adhesive tape, that tape will probably be much older than the tape on the new heater. The old tape will probably let go of the bed before the new tape lets go of the old heater. If the tape fails, the heater may get hot enough to burn itself up (thermal runaway protection in your firmware may help here).

If you're going to attack the temperature problem, why not attack the flatness/warping problem too?

[www.instructables.com]

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Along with from the usual precautions, you might want to swap out the wires for something more flexible. My Keenovo bed came with wires that are fairly coarsely stranded and I imagine they will fatigue and crack over time.

If you want to go mains, do so I found there were far fewer pitfalls than I expected. My setup has a circular 320mm 700w silicone heater, a SSR 25DA unit which doesn't get hot, its running only about 10% of its rated capacity, a RCD for the printer and a 150 deg C thermal fuse between the silicone and the bed. I have ensured the entire frame and bed are earthed.

Heat up time is 3-4 minutes to 120 deg C.

That being said my other machine has a 200x300 pcb heater which I run from a second PSU on 24v DC, but using the 12v circuit on the PCB. The PSU is only 250w so its not really a risk for the PCB (possibly the PSU but its well cooled), and an external mosfet board. Probably 8 minutes to 120 deg C so not too shabby.

I use plenty of insulation on both, I can wait for parts to cool before detaching them.

Edited 1 time(s). Last edit at 05/07/2017 03:40PM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
the_digital_dentist
If the existing heater is stuck to the aluminum using adhesive tape, that tape will probably be much older than the tape on the new heater. The old tape will probably let go of the bed before the new tape lets go of the old heater. If the tape fails, the heater may get hot enough to burn itself up (thermal runaway protection in your firmware may help here).

If you're going to attack the temperature problem, why not attack the flatness/warping problem too?

[www.instructables.com]

I was expecting your comments here!

The existing heater is bonded to the bottom of the ali plate, but I don't think it's with adhesive tape - I expect it's a glue bond. It's a fair comment about which glue would fail first though.

I have no idea whether I could get cast aluminium tooling plate in the UK, but I'd expect it to be expensive if it is available.

[www.clickmetal.co.uk]

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD
If you want to go mains, do so I found there were far fewer pitfalls than I expected. My setup has a circular 320mm 700w silicone heater, a SSR 25DA unit which doesn't get hot, its running only about 10% of its rated capacity, a RCD for the printer and a 150 deg C thermal fuse between the silicone and the bed. I have ensured the entire frame and bed are earthed.

Heat up time is 3-4 minutes to 120 deg C.

That being said my other machine has a 200x300 pcb heater which I run from a second PSU on 24v DC, but using the 12v circuit on the PCB. The PSU is only 250w so its not really a risk for the PCB (possibly the PSU but its well cooled), and an external mosfet board. Probably 8 minutes to 120 deg C so not too shabby.

I use plenty of insulation on both, I can wait for parts to cool before detaching them.

Sounds like the mains setup I was thinking about, apart from the thermal fuse - a very good idea.

My CoreXY runs on 24v using an external MOSFET board (both machines have an external MOSFET) but that still takes a while to get up to 100C. I don't fancy using the 12v circuit on 24v - I'm a coward!

I only do the 12v circuit as the 24v circuit was only drawing 4 amps directly connected to a 250w 24v PSU, which was not going to fly. If I had a 500w 24v PSU I wouldn't do it, instead I connected an external 24v fan to the mosfet controller in parallel with the heated bed, so that the fan ramps up as the PWM increases, so providing "titrated" cooling to the PSU, no issues so far been running it like this (with mech. relay until recently) for over a year.

Clickmetal make nice tooling plates very reasonably.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD
[www.clickmetal.co.uk]

Beat me to it! After my post I went away and did a search - found this supplier almost immediately.

So this is an option... hmmm....

Next question is - what sort of wattage should I consider for a 215mm square plate? (200mm square printing area, as on the common PCB printbeds). I don't want to be too extreme, just enough power to get up to around 110C in a reasonable time.

Generally accepted values are from 0.4-0.5W/cm². So a 200x200 bed would need 200w to have this power level. However having a little more available is not a problem if you are taking adequate safety precautions like the thermal fuse. You can always limit max PWM to the SSR (they work well on lower frequency PWM, or slow_PWM or whatever your firmware uses) if you want.

I got my thermal fuses from Maplin.

Edited 1 time(s). Last edit at 05/08/2017 06:40AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

The firmware for my Duet board can autotune the bed's PID, so hopefully that aspect would be covered. The automatically generated values can be modified after generation, and can be 'influenced' during the autotune process (I believe), if they don't work out properly.

I have a duet wifi so yes the process is automated I believe though you can limit max PWM. You wont need to at 200-300w. RRF will give you a dire warning if it thinks your heater is overpowered, but ignore it as it won't apply if you have a thermal fuse (in the sense that even if left on at full power the thermal fuse will disconnect the power below the ignition temperature of any materials nearby).

Edited 1 time(s). Last edit at 05/08/2017 10:23AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Thanks for all the input, everyone.

On reflection, I've wondering whether I should just go for a stick-on low-voltage heatpad for simplicity's sake - 12v on one machine, 24v on the other - at around 200W. They are available from aliexpress, plus a small number of UK suppliers (200W/24v may be difficult from the UK). I have an independent MOSFET device on each machine to drive the heatbed which, according to the maker's specs, will be more than capable of supplying the required current. I might have to add cooling fans to help them a little though.

200W should provide reasonable heat-up time for a 200x200 bed.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Have you considered this: [hackaday.io]? It's a PCB heater rather than an Al plate, but otherwise looks good. I have bought one, but have yet to set it up.

---

See my blog at [moosteria.blogspot.com].

Quote
animoose
Have you considered this: [hackaday.io]? It's a PCB heater rather than an Al plate, but otherwise looks good. I have bought one, but have yet to set it up.

That looks nice! It would be an exact replacement for my existing heater plate, and I see that it has a thermal fuse built-in - nice. I'm a bit nervous about those exposed connections though - I like my mains wiring and connections to be well hidden and untouchable...

Quote
animoose
Have you considered this: [hackaday.io]? It's a PCB heater rather than an Al plate, but otherwise looks good. I have bought one, but have yet to set it up.

At first glance, to me the safety implications of a mains powered PCB heater appear horrendous. Unless it has UL or equivalent certification (which I very much doubt), I wouldn't touch one with a (well-insulated) bargepole.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

That was my concern - I would want some serious insulation between me and mains voltage, especially on something I'd be touching frequently.