Heat mat voltage

I’ve been contemplating over and over again, reading and searching for information regarding heat mat voltage.
Basically I cannot decide what voltage heat mat I should use…

I intend to use cast alu plate (~200x200mm) with a silicon heat mat.
I want to have fast and stable heating, but Safety is a big concern.

I was almost dead set on using a 230v mains silicon heat mat in the 400w range.
But using 220v in a hobbyist built cantilevered design makes me nervous, even though I’ve done 230v installations before.
230v is dangerous stuff….

On the other hand, if I would go to 12v or more likely 24v, other issues arrives. Cable dimension, heat and so on as you will instead push “allot” of amp through your wiring.
Not to mention using 24v would mean you have to adopt all electronic to it, which by it self has it's ups and downs.

Lend me you insight and knowledge

400W is way over the top for that size bed heater. Most printers use 120W for a bed of that size. You could increase it to 180W if you want faster heating.

---

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].

So if you take DC's advice, 120W at 12V would mean 10Amps, at 24V only 5Amps. A quick google search will tell you what size wire to use but it won't be seriously thick in either case. If I had the choice, I'd opt for 12 or 24v just for safety reasons. 24v has other advantages and you could always use a 24 to 12v converter if your other electronics can't handle 24v.

10A @ 12V is also the limit of cheap electronics such as RAMPS, which is probably why Prusa i3 style printers generally use PCB bed heaters that draw nominally 10A @ 12v - although in practice the tolerances are very wide.

If your electronics can handle it, I would always choose 24V over 12V because of the additional flexibility it gives, e.g. higher power bed heaters without the current being too much for the electronics to switch directly, and stepper motor torque maintained to higher speeds (which is particularly important when using 0.9deg/step motors). Fans and heater cartridges that run from 24V are inexpensive and readily available.

---

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].

Thanks for your replays.

Yeah you are probably right, 400w is probably too much.
But certainly 120w can’t be enough… Especially considering I’ll be heating a 5-6mm thick alu plate with it.
120w is what? Your normal MK/PCB heatbed? Also known to take ages to heat up, if reaching the set temp (100-105c) at all…
And equals 0,3w/cm2 which below all recommended w/cm2 I could find.

DD is using 0,45w/cm2 to be heat his cast alu plate to 105c within 5min. So I’m using that as benchmark.

So I would not go below 200w, which would equal ~17A on 12v, or ~8A on 24v.
To me this rules out 12v. Which leaves 24v.
And I’m not entirely sold on why a 24v system (excl heatbed) is beneficial… And running a 12v system and 24v heat mat would involve some kind of relay/ssr whatever of which I haven’t found a solid/safe solution for yet.

I agree, with a 5-6mm alu plate then 120W isn't going to be enough. My delta has a 5mm alu plate and a power density of 0.4W/cm^2, which is a little high for safety because if left on at full power the temperature would exceed the 200C rating of the silicone heater.

Apart from making it easier to drive the bed heater, the other main advantage of 24V vs 12V is that motor torque is maintained to higher speeds. However, on a machine with a 200m square bed and using standard 1.8deg/step motors, the motors are unlikely to reach speeds where this makes a difference. A minor advantage of 24V is that you can use thinner wires from the electronics to the hot end heater.

A few people use dual power supplies: 24V for the bed heater, and 12V for everything else - usually because they are using 12V-only RAMPS electronics. Most electronics can be wired to use a separate supply for the bed heater, without having to use a SSR. Personally, I would much rather use one PSU for the whole printer, even if I didn't have a reason to prefer 24V for the motor drivers.

Edited 1 time(s). Last edit at 09/07/2016 05:32AM by dc42.

---

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].

Yeah, and that’s why I’m thinking of incorporating a thermal fuse to have some safety against thermal runaway of the headbed.
And why I prefer to over size vs under size, I don’t want to end up having to wait 15-20 min for heating.

That’s what I’m thinking. I can understand 0.9 degree more torque on deltas, but on a 200x200x200 Corexy I’m not sure it will have any effect.
Secondly I can’t find 0.9 degree 1.5-1.7A stepper motors that are not close to twice the price as the normal 1.8’s.

I have been thinking of running either dual PSU 12v+24v or 24v + buck converter down to 12v.
But in that scenario I need some kind or relay/SSR to switch the headbed on/off.
And I haven’t found a reliable source of what the best/safest option would be.
Probably a genuine DC-DC SSR, but boy they are expensive…

You don't need a SSR to drive the bed heater from obe voltage and the rest of the electronics from another, if the current draw of the bed heater is within the rating of the electronics.

btw a 120W PCB bed heater can heat up a 200x200mm bed to 100C reasonably fast (although not one made from 5mm thick aluminium). I have one that reaches 100C in about 10 minutes, and 120C if given long enough. The problems people encounter with these heaters may be caused by several factors:

- The resistance tolerance on cheap PCB bed heaters seems to be very large (probably due to variation in copper thickness). So a 120W nominal bed heater might in practice produce anywhere between e.g. 100W and 140W @ 12V.
- Most people who find these PCB heaters inadequate are using RAMPS electronics which uses a poor choice of bed heater mosfet, and (except for the most recent version) a polyfuse. Both of these components cause voltage drop in the bed heater circuit, which in turn reduces the power (e.g. 5% less voltage = about 10% less power)
- People often use cables that are too thin between the PSU, electronics and the bed heater, resulting in further voltage drop
- Some people use a cheap PC ATX power supply to power the 3D printer. These produce somewhat less than 12V on their 12V outputs when the 5V output is only lightly loaded, as is the case with 3D printers

---

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].

Well, in any case I'm not going to run the standard PCB heater + glass setup for many reason.
I rather go my intend way of silicon heater on a milled and casted alu plate with PEI.

Question really is how to do this considering different voltages in a safe, or safest way as this is of importance for me.

If I were to go 24v, I think the best way would be to go all 24v.
In that case I would probably need to go 0.9 degree steppers as the drivers I will use is DRV8825'a and they are known for their high pitch noise when paired with 1.8 steppers @ 24v.
I will run 32bit controller so it should be fine for 0.9 @ 1/32 steps.

Problem I find is the availability of 0.9 stepper and the price compare to 1.8.

Is there any down side to use 0.9 steppers? Should I aim for the same Amp 1.5-1.7A?

Quote
Amoniak
Well, in any case I'm not going to run the standard PCB heater + glass setup for many reason.
I rather go my intend way of silicon heater on a milled and casted alu plate with PEI.

I agree, it is a much better solution.

Quote
Amoniak
If I were to go 24v, I think the best way would be to go all 24v.

I agree. Except...

Quote
Amoniak
In that case I would probably need to go 0.9 degree steppers as the drivers I will use is DRV8825'a and they are known for their high pitch noise when paired with 1.8 steppers @ 24v.

The DRV8825 problem is not directly affected by whether the motors are 0.9 or 1.8deg, only by the motor inductance and supply voltage. So DRV8825 drivers are likely to be just as problematic with 0.9deg as with 1.8deg motors.

Best solution is to avoid DRV8825s. Here are some other options:

- Use traditional x16 drivers, after all x16 and 0.9deg should be equivalent to x32 and 1.8deg in terms of theoretical resolution, and better in practice because the positioning will be more precise. I am not sure how the noise compares. Changing my delta from 1.8 to 0.9deg motors made it quieter (using x16 microstepping), but I've never run it at 1.8deg and x32 for comparison.

- Use alternative drivers with >x16 microstepping. If you are looking to get a 32-bit controller, consider the Duet WiFi, which has TMC2660 drivers that do up to x256. Other drivers that do >x16 include the RAPS128.

Quote
Amoniak
Problem I find is the availability of 0.9 stepper and the price compare to 1.8.

Really? I find they are readily available, and only slightly more expensive. See e.g. [www.omc-stepperonline.com].

Quote
Amoniak
Is there any down side to use 0.9 steppers? Should I aim for the same Amp 1.5-1.7A?

Possible downsides are the slightly higher price, the need for faster electronics to generate the step pulses, and the need for 24V power if you are looking for really high speeds.

Yes, aim for the same amps, around 1.5-1.8A for most drivers, or 1.5 - 2.4A if you choose the Duet WiFi.

Edited 2 time(s). Last edit at 09/08/2016 12:08PM by dc42.

---

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].