Going from Prusa Heatbed MK2 to MK2b problem

My MK2 12v only heatbed shorted out and burnt, breaking the piece of glass I was using. I ordered a MK2b 12v/24v headbed from LearCNC and after replacing the wires it came with for decent ones I can only get it to 98 degrees (even without the glass). My PSU is at 11.88v with the bed on, the bed is at 11.7v with a 0.09v drop over the wire. The rest of the voltage drop (0.09v) is through the MOSFET, a 45a rated model. I'm using the same Thermistor, no firmware changes, wired on the board as per 12v instructions. Whats the deal? Aren't these newer models suppose to be better? I need to print ASAP and I can't with only 92 degrees on my glass.

BTW, the old MK2 would get to 112 degrees ok in around 9 minutes. I left this new one for almost half an hour.

Thanks in advance for any help and suggestions.

P.S. I've already asked LearCNC, waiting on a reply.

My guess is that your previous board tracks were too thin, and failed at some point, but could produce higher temperatures.

92 degrees is about normal at 12V. To improve you can increase the voltage if the psu has a potmeter, use insulation like cork under the heated bed or switch to 24V for the heated only.

Thanks for the quick reply. Back when I built this printer 4 years ago it didn't seem like anyone had the issue of 12v heatbeds not being able to get to 110 degree working temp. I did insulate it using the backing of a picture frame and a couple of layers of silicone baking tray. This helped speed the heating on the old bed up but it's not helping this bed. I'm using a PC 480w PSU, not sure if I can adjust it.

In theory it can be done but only if you're very comfortable hacking the psu. If you're considering a new psu, I'd recommend adding a 24V psu just for the bed. If you use a ramps board you can hook it up on the high current connector and solder the heated bed wires to use the 24V terminals. Heating will be much faster and the resistance from wires will not affect the heating much.

I don't know why exactly 12V heated beds don't reach 110 degrees that easy, my guess is that they've redesigned it to heat more evenly and lower the failure rate.

I have gen 7 electrics. I'm happy to try and make the PSU pump more through the 12v one but I don't want to cook my gen 7. Currently with no draw the 12v is at 12.4v I believe. What do I look for in the PSU?

I one want to spend much on this printer as I would just like to keep it as a backup once I build a new better one (when ever I get the time for that...)

I've found this on google:
atx psu higher voltage

This may be kind of a waste of money, a new power supply may work better for you. But I use these booster modules with the ATX power supplies for projects and testing makes it adjustable, but price isn't all that good. here is one on ebay.
LINK

I got mine from Amazon not sure if there is a difference other than I wanted it quick.

LINK

Hi,

Did you ever tryed to enable and adjust your PID values for the printed bed ?
I reach 115°C with my MK2Bs in about 8 minutes.
You can insulate the rear of your heabed too, thats helps a lot.

++JM

I have not adjusted any PID values but I can see and hear the fet is on continuously and the bed stops getting hotter at 93 degrees (with the glass on). As I said earlier the bed, both old and new, have the rear insulated.

I found a site with details on modifying ATX PSU's that use a three pin 431 IC as a voltage reference. Looks like I have to put a potentiometer in with the resistors on that IC's Ref pin then disable the over voltage circuit. It's a bit above my knowledge level but I'll give it a go.

Much less hassle to buy another cheap PSU for the bed, and run it through a SSR.

Quote
xelad
Much less hassle to buy another cheap PSU for the bed, and run it through a SSR.

If you do that, then unless it is a mains powered heat bed then you need to spend a lot of money on the SSR. The cheap DC-DC SSRs have a voltage drop of about 2V, which means they produce huge amounts of heat and therefore need massive heatsinks.

Unless you are comfortable modifying your ATX PSU to give more than 12V, then I suggest you buy a cheap 12V 25A LED power supply instead, and turn the voltage adjustment pot up a bit. A 10% increase in voltage gives you about a 20% increase in heater power. Two caveats:

1. You need to safeguard the mains terminals on the PSU, and provide a mains inlet connector or at least strain relief for the mains cable. I designed this enclosure for mine: [www.thingiverse.com]

2. If you are using Arduino-based electronics AND you have a LCD control panel connected, then increasing the supply voltage may cause the voltage regulator on the Arduino to overheat. You need to take appropriate mitigations measures, e.g. add an external 5V regulator, or replace the Arduino by a Taurino, or drive the L:CD backlight from 12V through a series resistor instead of from 5V.

---

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].

After some research I found a site that explained how to voltage mod an ATX PSU that seemed to line up with similar components to what mine had. (http://forum.allaboutcircuits.com/threads/variable-atx-power-supply.58707/) The TL431 they speak of was a KIA 431A on my PSU. There was 3 in total, one looked to be on the 3.3v line, the one I have labled in my photo had the resistors which connected to the 5 and 12v lines, not sure of the other one. I traced from the other side of the resistors to find which one was 12v, then found this diagram/datasheet of the WT7525 which gave me an idea of what it was (voltage and current protection). I figured that joining the poteniometer in series with the resistor would give the voltage adjustment as described in that other forums post. I read on the datasheet for the WT7525 that the fault protection latch ( FPOB ) pin was a drain, so it must have to drain current from something (a FET?) to keep running or the (FET?) will saturate? Anywho, I cut it's line and joined it to a ground so it is always "draining". This seem to disable the over voltage protection.

Without load the PSU is now 4.3V/14.4V with the pot turned to 0ohms, or 6.1V/17.2V with the pot turned to 1kohms. Being that the ATMEGA is only rated to 5.5v I disconnected what was joining the 5V output lines to the rest of the board, a thick coil, and put three 3A diodes in series to drop the voltage. Once back together I tested it under load and with the pot at 0ohm the volts were now 4.3V/14.5V and could be turned way up above 7V/23V, so I set the 5V back to 5V and the 12V line is now holding steady at 16.4V underload. The fan reacts a bit weird, higher the 12V line goes the slower it goes and makes some weird buzzing noises but at the 5V/16.4V setting it seems to be ok.

On this heat bed on the 12V connectors with 16.4V it now gets to 112 degrees in 9 minutes. Similar time frame as my old bed on 12.2V. At least it's working again and I don't have to buy anything else! I really need to print the parts out for my next printer just incase this one gives up on me again.

Thanks for the help people.

Edited 2 time(s). Last edit at 09/11/2015 09:39PM by Phizinza.

The good SSR's are running about $85.00.

And easier still to get a 24V and hook it up on the 11A connector and rewire the heated bed. No need for a SSR

Congrats on the PSU hack btw!

Edited 1 time(s). Last edit at 09/12/2015 03:02AM by imqqmi.

Quote
imqqmi
My guess is that your previous board tracks were too thin, and failed at some point, but could produce higher temperatures.

I was thinking about this yesterday, surely thinner would increase resistance and therefor decrease current flow??

Not if the length ie spacing of the tracks is likewise changed to keep the resistance the same.