12V heated bed with 24V to MKS GENv1.4

Okay, I want to run this past people to see if it's possible, or if I'm just crazy. I'm looking for the best use of the parts that I have.

My MBot 3D printer has an MKS Gen-Lv1.0 control board in it. It has working firmware (Marlin 1.1.0RC8) but it has some issues. The compiled firmware does not have provision for a heated bed, it shuts down the printer if the hot end goes to 230 deg C, which will sometimes happen before the printer will start (It hangs up on "Heating..." unless I use the controller to lower the hot end temperature setting. I can immediately set it back afterwards, but it's annoying, since I basically have to manually start the print after the hot end heats.) These are problems that new firmware should solve for me, but I did not want to overwrite the working firmware until I know that I've got a solution that works at least as well.

So I bought the MKS GENv1.4 board. (The GEN-L was out of stock at the supplier I bought from, or I'd have done that to have a direct swap.) The power supply that the printer came with is 24V/5A. I do not feel that it's a good idea to run the heated bed from that supply. I have a 24V/10A supply, but I want to use that for a second printer. I also have a 12V/25A supply from my old I3. I want to use this for the heated bed. Also in the equation is an external MOSFET board that I have in my bin o' parts. I also have a 12V 220mmX275mm heated bed. (I also have a 24V version of the same, but I want to use this one with a different printer, which will be controlled by the Duet Wifi board that I bought.)

So the plan is to connect the external MOSFET, supply it with 12V and connect it to my 12V heated bed. The rest of the printer will be run from the 24V supply, since things like the heater cartridge, motors, and fans are already configured for 24V, I don't see a problem with that.

My concern is the voltage mismatch between the control board and the heated bed, and how the actual resulting current will be affected. Since the control input to the MOSFET can go higher than the desired output, I know that I'll be able to get 100% output, and obviously I'm not going to get more than 100% power, but I may lose resolution in control. Funny, I can design an audio amplifier using MOSFETS, but this is out of my comfort zone. maybe I'm just overthinking it.

Possibility 1: Use the parts as I intended, because it will work fine.

Possibility 2: Use a voltage divider to halve the voltage signal to the MOSFET. Grab a couple of 4k7 ohm resistors and divide the output voltage in half. (4k7 should dissipate about 1/8W if fed 24V, so should be safe enough. 2k2 is getting a bit marginal for 1/4W devices.)

Possibility 3: Just give up and buy a bigger 24V supply, use the 24V heated bed, and buy (And wait 8-12 weeks for China Post) another one for the other printer. This is the most expensive option.

What would you go with, and why?

---

MBot3D Printer
MakerBot clone Kit from Amazon
Added heated bed.

Leadscrew self-built printer (in progress)
Duet Wifi, Precision Piezo parts

Your concern is about the Mosfets Gate voltage is higher than drain? It could cause issues, but I haven't studied any FET datasheet yet to confirm.

A workaround could be repurposing a PWM fan output jumpered to 12V ( if there's such a possiblity on MKS boards )

Edited 1 time(s). Last edit at 06/28/2018 11:50PM by o_lampe.

Yes, that is my concern.

I've been doing more reading on electronics forums, and have come to more or less the following conclusion:

Yes, it can be done, if you are using the MOSFET as a simple switch. When the gate voltage is raised beyond the drain voltage, it will act as a closed switch. This would be fine if the bed heater is configured as bang-bang.

Still, I think that this is probably at least somewhat problematic. I believe that I will be better off making a voltage divider, and working that way. It's a dirt simple circuit, will cost me practically nothing, even if I didn't have the components (which I do) and has very little risk associated with it. I suppose that there are some effects from parasitic inductance and the like, but this isn't all that sensitive a circuit.

So anyway, I guess I've answered my own question, and I'll use a voltage divider to tweak the input to the external MOSFET.

This does raise one more issue, and that's power supply control. I'd rather not have 2 switches for the power supplies, so I'll want to figure something out... But that can wait until the rest of it is working.

---

MBot3D Printer
MakerBot clone Kit from Amazon
Added heated bed.

Leadscrew self-built printer (in progress)
Duet Wifi, Precision Piezo parts

/facepalm.

You'd think that I'd look at the MOSFET documentation first? NO! Of COURSE not.

So apparently this whole thing works on an optocoupler and doesn't actually care what the input is. nope, I just overthought the whole thing. Of course the example that they gave is to power the bed at 24V and the mainboard at 12V, which is the opposite of what I'm doing, but looking at the actual schematic for my external MOSFET board, I can see that it doesn't actually matter, since the 2 circuits are never coupled, and the MOSFET is basically used as an on/off switch regardless.

Basically, it does mean that what I'm doing is, in fact weird. Probably at some point, I'll change over to just run the whole thing at 24V, but I'll run the heated bet at 12V for the time being, since that's the parts that I have.

Note to self. RTFM.

---

MBot3D Printer
MakerBot clone Kit from Amazon
Added heated bed.

Leadscrew self-built printer (in progress)
Duet Wifi, Precision Piezo parts

Just to round it up: You could have used PID controlled PWM mode too, because a MOSFET always acts as a switch, but in PWM mode it is switched on/off with varying duty cycle pretty fast.

Well, a MOSFET can be used to act as a valve, though the RAMPS hardware doesn't. The external MOSFET really doesn't, now that I've done a more thorough analysis of the circuit. It really is configured to act like a relay, but without the pesky arc faults at the contacts. So yeah, I have it configured in PID mode, and it works perfectly.

The following is rambling that you probably don't need to read. The important part is above. It works.

My power supplies look a little goofy, since I have 2 of them. Somewhere along the way I blew the fuse in my multimeter. No idea how, but it's a $5 Chinese meter, so who knows.

Everything works okay, but the motors are being really noisy. I obviously have to tune my A4988 stepper drivers, so I'll have to find a fuse for that meter. I'm tempted to bypass the fuse with a single strand of copper wire. That would be more than the 250mA that the fuse is supposed to be for, but probably not more than about 700mA or so before it popped. It should be okay if I don't do anything dumb with the meter. Probably would do permanent damage to the inside of the meter if I do something dumb. It galls me though that the cost of a pack of replacement fuses is >80% of the cost of replacing the whole meter. Maybe I can find one of the old ones that I junked when the probe got broken, and scavenge a fuse from it. Or scavenge the probe from this one, I think the one in the junk pile is an auto-ranging digital, this is manual select analogue. Well, probable worst case with the single strand bypass is that I wreck the meter.

I've replaced the MKS control board, and I'm using a different set of A4988 drivers. The board that came with had the current set, and the motors were fairly quiet. These drivers were set for a different situation. They seem to work, though the extruder makes noise, and seems to under-extrude, even given the same steps/mm setting and same gcode file as worked before on this printer.

---

MBot3D Printer
MakerBot clone Kit from Amazon
Added heated bed.

Leadscrew self-built printer (in progress)
Duet Wifi, Precision Piezo parts