Welcome to the SeeMeCNC Community

Personally think 24v is a bit excessive. Todays plan is to hook the bed up to a variable psu and wind it up until the bed equalises at about 130 maybe 140. I reckon this will be about 14 - 15v. It might not warm up as fast as higher voltage but if the driving ssr fails its a built in limiter. I reckon 19v would hit thermal equilibrium at over 200c (assuming 20C ambient) great for cooking baked potatoes but not so good for your printer.

There's two reasons I like 24Vdc. First is that power supplies are readily available, especially if you work with a bunch of industrial electricians. Second is that I can use a control relay with a 24Vdc coil, and not have to worry about having the 12V trip it somehow. There are few enough relays that will like variable voltages in the first place, and finding one that would trip at 14V but not at 12 is hard.

As for built in limiters, well, add a second relay with a low voltage coil, and drive it directly off the main power supply (If you're using the ATX supply, use the 5Vsb rail so as soon as you plug in it goes live, otherwise use the 12V from the PSU (You could use wall voltage, but I don't want any more AC wiring than I need inside my printer, given the unshielded endstop wires)) . Make it a 4PDT (Since they don't make a whole lot of ST relays with that many contacts, and you can ignore the other sets of contacts) Use the NO contacts with a thermal fuse on the line to the coil, say a 235C fuse seated under the bed. That way if your SSR gets stuck, it fails in a predictable fashion. But if you don't have a way to interrupt the 24V PSU's wall voltage, you need to use a 60$ mechanical relay. (Say, have a 3.5MM jack that goes with any power in, that has the sole purpose of carrying signal from your master relay to another relay in the power supply unit (My god, my PSU brick design keeps getting bigger. Circuit breaker, switch, IEC socket, Relay, 3 connectors now)) At which point you can use a pair of 5$ relays (And rarely replace them, since they should only switch when turned on and off, and thus it should be under minimal load, giving them a very long switching life)

Realistically, there's no need to head the bed any higher than 80c for ABS prints.

g.

A couple of points:

1. IMO there are two main advantages of using 24V power instead of 12V, which may or may not be relevant to your machine:

- You can have double the bed heating power before you need to use an SSR

- You can get higher speed from the motors. This is especially relevant when using 0.9deg/step motors on delta 3D printers. Probably less relevant in a CNC machine.

2. Maximum heated bed current for direct drive from the Duet:

The Duet 0.6 is happy at 10A bed current and probably OK at 12A. If you go much higher than that, the PCB trace at the edge of the board on the top side that connects the Vin+ and bed heater+ terminal block pins may burn out. You can forestall this by linking those two pins with a thick piece of wire on the bottom of the board. As noted earlier, the bed heater terminal block is a rather small for more than 10A.

The Duet 0.85 has a larger terminal block for the bed heater and a thicker PCB trace, so should be OK at a little over 12A. Again, you can link the Vin+ and bed heater+ terminals on the back of the board if you want. Note that the + and - bed heater terminals are the opposite way round on the 0.8.5 board compared to the 0.6.

dc42 wrote:A couple of points:

1. IMO there are two main advantages of using 24V power instead of 12V, which may or may not be relevant to your machine:

- You can have double the bed heating power before you need to use an SSR

In case anyone reads this and might misinterpret it, this assumes you're sizing the entire circuit for this purpose (i.e., using a higher-resistance bed). Doubling voltage on a stock Onyx would also double the current. The resistance on my Onyx was low enough I was barely comfortable running it without an SSR at 12V (a situation I have since remedied).

Anyone got recommendations for a 24v fan/duct?

dc42 wrote:A couple of points:

1. IMO there are two main advantages of using 24V power instead of 12V, which may or may not be relevant to your machine:

- You can have double the bed heating power before you need to use an SSR

- You can get higher speed from the motors. This is especially relevant when using 0.9deg/step motors on delta 3D printers. Probably less relevant in a CNC machine.

2. Maximum heated bed current for direct drive from the Duet:

The Duet 0.6 is happy at 10A bed current and probably OK at 12A. If you go much higher than that, the PCB trace at the edge of the board on the top side that connects the Vin+ and bed heater+ terminal block pins may burn out. You can forestall this by linking those two pins with a thick piece of wire on the bottom of the board. As noted earlier, the bed heater terminal block is a rather small for more than 10A.

The Duet 0.85 has a larger terminal block for the bed heater and a thicker PCB trace, so should be OK at a little over 12A. Again, you can link the Vin+ and bed heater+ terminals on the back of the board if you want. Note that the + and - bed heater terminals are the opposite way round on the 0.8.5 board compared to the 0.6.

Thanks for posting this info! I was just wondering about my current build.
So what is minimum acceptable resistance of heated bed at 24V without going with SSR?
I have a Onyx rev.3 that measures 1.2 ohms.
I would really prefer to run my Duet 0.85 without SSR, for simplicity.
Also, I could adjust the psu to a bit less than 24V, not sure how low though, 20V maybe?

bob64 wrote:Anyone got recommendations for a 24v fan/duct?

wire two 12V fans in series

Brian, I know that we all want simplicity but as a former electronics technician, I have to agree with MHackney.
The traces on the Duet and all cards I have seen are not built with a thick enough or wide enough traces to carry the approximate 20 Amps going to a heatbed
and leave a margin of safety. Simplicity over safety loses everytime. I personally use the SSR and a 24VDC to 12 VDC converter so that the bed is the only thing
running at 24VDC. Each of you own your printers and can do whatever you want with your printer but I say it again, these present day controllers were not made
to handle the kind of wattage that 24VDC to the bed requires. Happy Fires to all!

bob64 wrote:Anyone got recommendations for a 24v fan/duct?

24V fans are widely available on eBay as well as from the usual component distributors.

If the 0.85 is good to 12A as stated, at 1.2 ohms you shouldn't put more than V = I R = (12A) (1.2 Ohm) = 14.4V through it. I haven't tried to get a 24V supply down that low. Some overdrive 12V supplies to around that voltage, but I wouldn't.

bvandiepenbos wrote: Thanks for posting this info! I was just wondering about my current build.
So what is minimum acceptable resistance of heated bed at 24V without going with SSR?
I have a Onyx rev.3 that measures 1.2 ohms.
I would really prefer to run my Duet 0.85 without SSR, for simplicity.
Also, I could adjust the psu to a bit less than 24V, not sure how low though, 20V maybe?

If your bed heater measures 1.2.ohms then it would draw 20A @24V which is far too much for the Duet. Also the heating power of 480W would be rather high for a bed that I think is about 300mm diameter (the product page I found for it didn't give it's diameter). You have a couple of options:

1. Use a 12V supply turned up to 14V, or 15V if it will go that high. You can put two silicon diodes in series with your fans so that they only get about 12V.

2. Use a 24V supply turned down to about 20V and a low voltage drop SSR.

1.2 ohms is a rather awkward resistance for this size of bed heater. The ideal value for.driving direct from a Duet at 24V would be about 2 ohms.

It took me long enough but I finally got the upgrade working.

Thanks everyone for your help! I really could not have finished the upgrade without your all's help.