heatbed won't turn off

[dwill207]

I just installed a printrbot heatbed, rev F printrboard, and atx power supply. Without even having the printer connected to my computer, the heatbed became hot right away. Is this normal? It seems like I should be able to switch the heatbed on in my slicer and not have it come on automatically. Once I connected the printer to my computer, I opened my slicer and it showed the bed temp was not registering. No matter what I did with the slicer, no change was noticed. Is my new rev F board faulty from the factory?

Thanks,
Duane

[frankv]

[RetireeJay]

Just a little check here... Take a look at the components with the metal tabs that are sticking up from the Printrboard right next to the connectors for the heated bed and the hot end (the components are MOSFET transistors in TO-220 packages). If the metal tabs of the two transistors are touching each other, bend the transistors gently (one forward, one back) so the tabs don't touch and try again. Also, look for any signs of a short circuit. The transistor is supposed to be "off" when you turn on the board. When it goes "on" then it grounds one lead of the circuit to the heated bed; the other lead is connected directly to +12V. So anything that bypasses the transistor and connects the switched lead directly to ground will make your heated bed be "on" all the time.

When you say the heated bed temperature is not registering, exactly what do you see? Zero degrees? 300 degrees? or room temperature? There are different solutions depending on which case you have.

[dwill207]

I checked the MOSFETs and they aren't touching. Also, I can't find any visual signs of a short. I don't have any way to test the transistors. My slicer is showing the bed temp at 27 C. It's never gone over 27 even when the bed was hot. I've since disconnected the heatbed so I can use the printer without it. I'm fearing one of the MOSFETs might be bad. Would it be one or both?

[RetireeJay]

Well, we have two separate problems here.

The readout of bed temperature being 27C means that you do have a thermistor connected, and the board "sees" it correctly, but the thermistor is not "seeing" the heat from the bed. So it seems like an installation problem: the bed thermistor needs to be physically attached to the bed either with tape or with some mechanical means so that the thermistor's temperature is equal to the bed's temperature. For some reason, your thermistor is just reading ambient room temperature. (One possibility is that you have the thermistor for the Hot End and the one for the Heated Bed swapped - but you'd see the Hot End temperature going up instead of the Heated Bed when you connect the printer.)

The MOSFET that is closer to the power plug on the Printrboard is the one that controls the Hot End. If your Hot End doesn't heat up prematurely, and if it is controllable, then there's no reason to suspect that one.

If you have a voltmeter and a bit of electronics expertise, you could look up the schematic for the Printrboard (on RepRap.org/wiki) and try looking at the gate voltage on the transistor. If it's always at +5V, then the transistor is probably just doing what it's told to be doing - turning on. If that's the case, then MAYBE re-flashing the firmware would help. I don't often recommend re-flashing, but in this case you have nothing to lose by trying; your board is not usable for the heated bed. (If the gate voltage on the MOSFET transistor is zero but the transistor is a short circuit all the time from source to drain then you could replace the transistor.)

[dwill207]

Thanks for the help, but this is over my head. I do have a voltmeter but don't know which terminals to check and can't find a schematic. I also tried to remove a MOSFET from my old revD and it wouldn't budge. So, seeing as I couldn't remove one from my old board, there's no way I could remove it from my new board without damaging it. I guess I'll just wait to see what Printrbot says. Maybe they'll let me exchange it...? I won't hold my breath though. If I do have to purchase another rev F, I might have the same problem. There's no way to make sure Printrbot sends a good board is there... : /

[RetireeJay]

The MOSFETs are soldered in, so you need a soldering iron (or gun) and preferably some sort of solder-sucker or solder-removing copper braid to "unsolder" the transistors so you can swap them. But before you do that, let's look at alternatives.

First use your multimeter. Here is a picture of the corner of the board layout. This picture is oriented looking from the top of the board, but you'll have access to the pins from the bottom, so remember that this is, in effect, an X-ray view. I've numbered the terminals on the MOSFET for the heated bed 1, 2, and 3 in red. Terminal 1 is the control terminal, or "gate." Terminals 2 and 3 are the "source" and "drain." With the power off, measure the resistance between pins 2 and 3 with your ohmmeter. Measure in both directions. If, in both directions, you get zero ohms (or just one or two tenths of an ohm) then the MOSFET is bad. If you see a significantly higher resistance (in either direction) then the MOSFET is probably good.

Then, if you can do so safely, turn on the power and measure the voltage (not the resistance) between pins 1 and 3. If the voltage is zero (at turn-on, without any programming) then the transistor SHOULD be "off" and pulling no current. If the voltage is +5V, then for some reason the microprocessor is telling it to turn on (or there's another defect in the board physically) In that case, re-flashing is definitely in order.
Printrboard Rev F Bed MOSFET.JPG

BTW, here is the link directly to the RepRap.org schematic and board layout: (I don't know why you couldn't find it.)

[dwill207]

Thanks... You didn't have to do all that for me but I appreciate it. I am aware that a soldering gun and solder sucker is needed to remove the MOSFET. That's what I was trying to do. It didn't work... I wasn't able to remove all of the solder. So, I measured the resistance (with my ohmmeter set at 200) between 2 and 3 and I was getting readings between 10 and 25. Then I measured the voltage between 1 and 3 with the heatbed set to "off" and there wasn't any voltage. I also measured with the heatbed set to "on" and I saw 4.95 volts between 2 and 3. It seems like the MOSFET is ok. However, there is still 12 volts at the heatbed connection terminal whether it's set to "on" or "off". I re-flashed my board and I'm getting the same result, (12 volts at the heatbed terminal). I'm going to attempt to return the board to Printrbot and order another one.

[Mooselake]

There's always 12V at the heatbed terminal, relative to ground. The bed is controlled on the ground side, open for off (so the circuit isn't completed), closed for on. You can see in the schematic.

If the mosfet is good, and being switched properly, then consider a wiring error or a solder bridge somewhere else between the trace from the mosfet to the connector.

Kirk

[dwill207]

Thanks Kirk... I had a very hard time understanding that. I swore if there was voltage there, the bed should be on. I'm still baffled. It's obvious I don't know much about circuitry. While there WAS a problem when I first installed the heatbed, it was probably remedied after I re-flashed the board the first time. I just didn't know it. Your comment made me reconnect the heatbed, just to see.... and it works now! Lesson learned.

[Mooselake]

Printrbot does have an obligation to deliver what they sold (it's called merchantability and fitness). Open a ticket for your DOA board

Kirk