Fitting a 24V PCB heat bed?

I’ve been thinking about ways to increase the heating capacity of my bed and have bought one of the MK2B dual power 12V/24V PCB heatbeds from Hobby Components to try out. I am thinking I can use the 12V bed feed from my Melzi to switch a 24V PSU via a relay instead of powering the bed directly. Will the Melzi still be able to regulate the bed temps with this setup? Obviously I'd recalibrate the PIDs.

The new bed is rated 120W at 24V, so a 10A (or even 7.5A) 24V laptop brick should have plenty of headroom and a 12V relay is easy to source. Am I missing something obvious?

Edited 1 time(s). Last edit at 11/27/2014 11:07AM by QuackingPlums.

Yes you connect the two halves in series for 24V so it takes 5A and gives 120W. For 12V you connect them in parallel and it takes 10A and still gives 120W.

---

[www.hydraraptor.blogspot.com]

@QuackingPlums: I do not believe you gain any extra heating ability by running the MK2b at 24V, it just permits you to use smaller gauge wires.

You can achieve a better effect by running a 12V bed at e.g. 18V.

For the 32-bit controller I am presently designing, I am using a switching regulator to generate 5V so I will be able to tolerate input voltages up to 36V to gain extra heating from a 24V bed.

Regards,
Neil Darlow

---

I try to write with consideration for all nationalities. Please let me know if something is unclear.
Printing with Mendel90 from fedora 25 using Cura, FreeCAD, MeshLab, OpenSCAD, Skeinforge and Slic3r tools.

I run an MK2 bed at 17V with an adjustable power supply to get 130C on the surface of the glass. You can use the existing MOSFET on the Melzi you just have to make a solid ground connection from the second PSU to the ground on the Melzi and feed the bed + directly from the 17V PSU. The MOSFET on the Melzi just completes the connection to ground.

When the bed is cold it takes 17A but when it warms up to ABS temps it drops to about 10A. I think the MOSFET on the Melzi will handle that without a heatsink but you may have problems with the Melzi ground terminal rating.

---

[www.hydraraptor.blogspot.com]

Ah, I've misunderstood how the 24V bed works then. I was assuming that I need a bed to be capable of running at 24V in order to get away with increasing the bed supply voltage. It sounds like the 24V bed just reduces the need for a high supply current without actually increasing the bed temperature. As the Melzi happily works with the 12V/10A draw of the current bed it looks like there is no benefit to me in switching to this 24V bed right now, especially if the 12V bed is capable of being driven at 18V.

My 12V bed currently manages good adhesion temperature only within a 100x100(ish) square about the centre point. At the edges this drops to almost Tg.
If I understand these instructions correctly then I just need to find a 17V-18V power supply and bond the ground line of this to the existing ATX PSU, and then drive the existing 12V bed from the new PSU?
The ground terminal issue you mention - is this the big green one where the black wires come in from the PSU? Any reason why I shouldn't take the ground directly from the bundle of black wires, whilst keeping as short a bundle into the terminal as I can?

Quote
QuackingPlums
If I understand these instructions correctly then I just need to find a 17V-18V power supply and bond the ground line of this to the existing ATX PSU, and then drive the existing 12V bed from the new PSU?

Yes

Quote
QuackingPlums
The ground terminal issue you mention - is this the big green one where the black wires come in from the PSU? Any reason why I shouldn't take the ground directly from the bundle of black wires, whilst keeping as short a bundle into the terminal as I can?

Yes you need to make a T junction near to the Melzi terminal to common the grounds but all the bed current will still flow through the Melzi ground terminal. With the bed current increased from about 10A to 17A when it is cold it may exceed the rating of the terminal. Possibly you can fit a bigger one. In either case it needs to be a very good low resistance connection.

It may not help your problem as if you make the edges hotter the centre may be too hot. Possibly an aluminium bed might be a better investment.

Edited 1 time(s). Last edit at 11/29/2014 12:04PM by nophead.

---

[www.hydraraptor.blogspot.com]

Quote
nophead
Yes you need to make a T junction near to the Melzi terminal to common the grounds but all the bed current will still flow through the Melzi ground terminal. With the bed current increased from about 10A to 17A when it is cold it may exceed the rating of the terminal. Possibly you can fit a bigger one. In either case it needs to be a very good low resistance connection.

It may not help your problem as if you make the edges hotter the centre may be too hot. Possibly an aluminium bed might be a better investment.

Ah yes of course. I was only thinking about the resistance of the connection. I'll investigate higher rating terminals but I did wonder what the effect would be of having the centre so much hotter.
Is it worth looking at other means of spreading the heat more evenly across the PCB bed or am I just wasting time chasing that one?

So i can change the 12v of the bed to a 17v adjustable power supply and heat the heatbed by hand?

i have an A4 sized aluminium plate on an underpowered MK2A heatbed, and i need to up the power on it.

Where can i find/make a 17v adjustable power supply that provides 10-15 amps? can i just hook up a regulator to an 18v power supply?

I use one of these: [www.ebay.co.uk]

I still control the bed with the Melzi, no need for manual control.

@QuackingPlums,
Sorry didn't notice your last question. A 1mm thick sheet of aluminium between the PCB and the glass seems to help spread the heat.

---

[www.hydraraptor.blogspot.com]

hm... using 24V with the MK2a would be too much I guess? Because 24V power supplies are much easier to get than 17-20V power supplies.

I think 15V is about the max I would recommend on the MK2a we supply. They are lower resistance than MK2b mentioned above.

24V would be four times the power and exceed the current rating of the connectors and wiring. At some power it will cause the tracks to de-laminate from the PCB. Also it won't max out at a safe temperature in a fault condition.

I think you mentioned you have an aluminium bed? What temperature are you trying to achieve?

---

[www.hydraraptor.blogspot.com]

It is not so much about the maximum temperature, I managed to measure 108°C on the surface with an IR thermometer and ABS stuck to the bed just fine so that is not the problem ( a little more would be nice too)... but it is a little annoying to wait 30 minutes to reach that temperature.
For PETG I use 90°C bed temperature and that takes around 5 minutes to reach (haven´t really measured yet) and I was hoping to speed up the heating process.

I was planning on using something like that [www.ebay.de]

I don't know why it takes 30 minutes. I have a 4mm aluminium bed on one of my Mendel90s with the stock PSU and it gets to 110C much faster than that. It maxes out at about 120C and does take a long time to get there unless I cover it with a sheet of cardboard while it warms up. The cardboard has a slot for the nozzle at the front. However I found I didn't need 120C for the first, 110C works fine on PEI coating.

I have been meaning to design a folding bed cover that is hinged at the back of the machine with a weight to make it swing up vertical and fold double on itself. To start the build I would swing it down and fold it out over the bed, clipping it to the front. As soon as the bed moves forward the clip would let go and it would swing out of the way. Or alternatively a roll of cloth could roll up like a blind with a little weight on a pulley, or a spiral spring.

I don't think that power expander is any use because the MOSFET is not as good as the ones on the Melzi and the connectors look smaller as well.

---

[www.hydraraptor.blogspot.com]

The power expander I´d use because it seems to be the easiest and safest way to do it (for me). Using the Melzi, I´d probably end up frying the whole thing. You don´t think that using a 15V power supply with the power expander and the MK2a would get me the desired heat up boost?

I´ve read about covering the build plate during heat-up, but since I like your method of putting the hotend onto the bed to prevent oozing, I don´t use it. In fact I did try during my ABS test run... but it still took almost half an hour to heat up to nearly 120°C (thermistor value) and when I removed the cardboard it instantly dropped again.
Maybe I should think about better insulation on the underside of the MK2a? So far, I have the cardboard sheets you described in the manual under it, maybe there is a better method?

But making a cutout for the hotend is a good idea, will try that. Don´t know If I can remove it with the hotend down though, my chamber might get in the way.

15V would give 50% more power, so it would heat a lot faster because. Not only is it rising 50% faster, it is also heading to a higher temperature, so doesn't level off as much before it reaches the target.

The power expander won't help it heat any quicker as it will have a slightly bigger voltage drop than the MOSFET on the Melzi.

If you use a 15V supply you only need to connect it to the bed + wire and the Melzi ground, so other than putting a bit more current through the Melzi ground terminal it doesn't affect anything else. You do need to make a decent Y junction in the ground though.

The problem with manually removing a cover is you have to be watching over it else it causes a head crash. Maybe you could put a pause command in the gcode that waits for a host button press.

Fibreglass loft insulation or ceramic wool kiln insulation will be a better insulator than corrugated cardboard but I don't like handling it and with it moving it might shake off fibres. I would probably wrap it in Kapton or PET to keep them contained.

If you do use a 15V supply my preferred bed material for ABS is glass with very very very dilute ABS juice baked on at 148C (surface temp ~130C) for the first layer. I can then print on it at 125C which gives a surface temperature of about 105C.

Edited 2 time(s). Last edit at 03/25/2015 06:14PM by nophead.

---

[www.hydraraptor.blogspot.com]

Quote
nophead
If you use a 15V supply you only need to connect it to the bed + wire and the Melzi ground, so other than putting a bit more current through the Melzi ground terminal it doesn't affect anything else. You do need to make a decent Y junction in the ground though.

That doesn´t sound too difficult, maybe even I could manage without frying anything, but just to be sure....

So I remove the + connection of the bed to the melzi and connect it to the 15V powersupply, the ground connection of the bed stays connected to the melzi but the GND of the 15V power supply I connect to the Melzi ground as well?

Yes, exactly that.

---

[www.hydraraptor.blogspot.com]

Thank you, I have added some LED strips to light the interior of my chamber, so far I have used a second 12V power supply to power those (was planning on building an chamber heater as well, that is why I didn´t use the original power supply in the first place), but since supplying the bed with a third power supply seems to be a little excessive, I guess I could supply my lights through the original power supply or is that not a good idea (using the 15V power supply for the bed should free some juice for the lights)

I can't see a problem with that. Some LEDs flicker when the bed switches but if you put that on another supply it should fix the issue of the 12V rail changing.

---

[www.hydraraptor.blogspot.com]

The 15V power supply arrived today and I connected the + Wire of the bed directly to the 15V power supply and for the GND connection of the bed I made a Y-Connection so that it is now connected to both the GND of the 15V power supply and to the GND of the Melzi Hot Bed Connector.

Now the Bed starts heating immediately after I switch on the 15V power supply even though the temp setting for the Bed is off. Did I wire something wrong? How do I get it to respond to the Melzis control?

Your description of the wiring looks correct but it shouldn't behave like that so all I can assume is you made a mistake implementing it.

---

[www.hydraraptor.blogspot.com]

Quote
D4RK1
for the GND connection of the bed I made a Y-Connection so that it is now connected to both the GND of the 15V power supply and to the GND of the Melzi Hot Bed Connector.

That sounds like your bed is always on as it's always connected to the 15V power supply. You need to have the ground connection of the bed go to your Melzi, and then from your electronics to the ground of the 15V supply. Two supplies of different voltages can share that same connection.

Ah, yes I didn't spot that. The Y connection should be in the ground of the power connector of the Melzi, not the bed ground. I.e. both power supply grounds go to Melzi ground. The bed ground is switched by the Melzi and bed positive fed from 15V.

---

[www.hydraraptor.blogspot.com]

ah... today I made a diagram of my wiring and I thought "Ok, it is obvious why it isn´t working" but lacking real understanding of electronics I didn´t know how to do it right.

But I think I got it this time, will solder the 15V GND to the Big bunch of wires (black ones) that come from the 12V PSU. The + wire from the bed will still go directly to the 15V PSU and the GND wire of the bed will go to the Hot Bed GND Connector of the Melzi.The Hot Bed + Connector of the Melzi will stay empty.

Thanks, Jens

Success,... it´s working as it should be. Thanks again for your help!