CR 10S Pro heated bed runaway

I have a CR 10S Pro that begins heating the bed immediately when turned on. I was printing and smelled something like insulation burning. Then I noticed the temperature of the bed climbing beyond the set point. After a while I took the base off and found the SSR for the bed was broken away from the base. I purchased a new SSR, installed it, turned on the printer and the bed still heats up on its own. I have taken videos of what is going on and sent them to Creality. The answer from them and others is a failed MOSFET that controls the SSR on and off.
I am posting this because I had trouble finding help for this. Alden Cost has a video on YouTube called “Creality CRX/CR10S Pro SSR goes up in smoke digging into the problem, coming up with solutions!”; however, this only covers the SSR, not what causes the runaway.
This is a potential fire hazard and needs to be fixed by Creality. Thermal runaway.

2 Likes

Here is the link to the YouTube video. https://youtu.be/C69VQQ6HvF0

1 Like

Here is a picture of the SSR.

1 Like

Disconnect the control wire from the Solid State Relay. It’s the small red wire. Turn the printer on and verify that bed does not heat up. We’re just confirming that the SSR is not defective. I hope you kept the old SSR. You should test it before throwing it out.

Is the printer still under warranty? If so, lean on Creality a bit and make it their problem.

The SSR is itself a large MOSFET. The control signal for could be a smaller MOSFET, a transistor or op-amp (operational amplifier).

Before we go further, verify that you are out of warranty as trying to repair it yourself will certainly void any warranty you still have.

Are you competent with a multi-meter and surface-mount soldering skills?
If not, this is going to require replacing the motherboard, if so, send a close-up photo of the controller board.

1 Like

I think their recent firmware has thermal runaway protection. If not tiny machines does, id suggest updateing the firmware while your at it. Teaching tech has a video of how to update the firmware.

Thanks. I didn’t tell all I did. Yes I checked the SSR and the signal coming from the MOSFET. Creality is sending me a new MB. Also, I have ordered one if theirs doesn’t arrive soon.

1 Like

I had Tiny Machines V 6 running at the time and it is supposed to have thermal runaway. It didn’t do anything.

1 Like

Weird. Well other than following the smoking wires, which you seem to be doing, i dont have much more to add. Though i would like to see how this progresses as i have a cr10s pro aswell.

Thermal runaway protection is irrelevant and likely not even active in this case.

Thermal runaway protection is a software system wherein the software checks the beds reported temperature and compares it to what the firmware is expecting it to be. Hence, if the firmware is calling for heat but the bed isn’t heating (possibly due to a bad thermistor), it shuts the heater down. If the firmware is calling for a given temperature, but that temperature is exceeded beyond some safety margin, it shuts the heater down.

In this case, however, the firmware wasn’t calling for anything. It had no expectations for a target temperature and hence, nothing to compare with, and couldn’t shut it down even if it did detect it.

1 Like

I understand that for when the printer was first turned on. When the SSR failed, I was printing and the bed temperature exceeded my setting. When I put the new SSR in and turned on the printer, the light on the SSR turned on immediately. That suggested to me that the relay is normally closed and the MOSFET opens the “contacts”. If that is the case, 24 volts are coming from the power supply directly to the bed and the MB could try to turn it off, but the SSR can’t.

This is NOT something I wanted to read, considering I bought a Creality CR10 Max last September. As big printers are usually bought to print big objects, and big objects can take 10’s of hours, or days, to print, this is a serious potential problem. Every time I turn the Max on, I’ll be wondering if its at all safe.

To be fair, I’ve already done several prints with my Max that lasted overnight, and now I wonder if I’ve just been lucky (so far), or stupid. I’ve never seen evidence of what you describe with the SSR and MOSFET, or bed overheating beyond my setpoint, but there must be a way to check for this kind of problem, short of waiting for the printer to set my house on fire.

Please keep us apprised of what you learn. I’m glad to hear you caught this before disaster struck.

Sounds like to me it is the mosfit on the board and it’s stuck open so it’s constantly heating you need to replace your main board

Actually, what it suggests is that the controller board’s MOSFET/Transistor/OpAmp/whatever, has an internal short that is immediately connecting the external MOSFET/SSR’s gate (control input) to 5V (or whatever the control voltage is), thus turning it on. Your SSR is behaving normally; it’s the control board that has a problem.

I am preparing a document with photos of what I know. I’m trying to understand how this works devoid of any documentation. As soon as it is complete, I’ll post it for yours and others input.

@f11 @ajordan
There is a simple solution to the concerns this brings up about thermal runaway. Visit this Amazon store and select a temperature above where you think you will ever be printing. Glue or tape one of these fuses to the underside of your bed and wire the fuse in series with the SSR control input. If the controller ever fails in the way ajordan’s did, this will kill power to the relay. Frankly, this will work for any printer that doesn’t have thermal runaway protection either because it’s not in the firmware or the hardware has failed.

Thx for the link !!

As an external (ie non-printer fw/sw controlled) fail-safe, that might work … hmm, one on the underside of bed as you say, perhaps another attached to the hot-end. Wire both in series to input sense line of external N.O. relay in series with printer or heater power - if either fuse goes, relay drops out and opens power to printer or heater. Kinda brute force, and definitely ugly but … I’d prefer something from Creality built into the heater/controller circuit.

If this is a common problem (not saying it isn’t a one-off), then I’d think CSA Approval as an electronic consumer product might be up for a review. If its a one-off, I guess its a user problem.

But I like the belt-and-suspenders workaround you suggest.

It’s not quite as simple as you’re thinking. Yes, you could put a spare Solid State Relay in series with the printer’s power input, and that will kill power to the printer if either thermal fuse blows, BUT, how would you turn the printer on? Remember, the series SSR would need power of it’s own first, in order to provide power to the printer. That’s why I suggested using it to disable the SSR for the bed. You could put it in series with the bed heating element, which would also work, but as I recall, those tiny thermal fuses I linked to don’t have a very high current capability.

For what it’s worth, I got the new board today and finished my analysis. Here is what I learned:

CR 10S Pro SSR Failure

Here are the things I know about this problem:

THIS IS WITH OLD BOARD

    1. The SSR is normally open by the schematic on the SSR.
    1. When I check the output of MB to the input the SSR, I get 24.22 V from my Fluke 77 multimeter.
    1. With the + on the SSR connected to V+ on the power supply and the – on the SSR going to the bed and I check from the + terminal on the SSR to the V- on the power supply, I get approx. 0 V.
    1. With the old MB, when I turn on the printer, I get 24.22 V across the MOSFET on the MB. This is the input to the SSR. Therefore, this would enable the SSR.
    1. However, when I check across the output terminals of the SSR, I get 24.22 V, but the bed is not heating.
    1. If I connect the input terminals to the terminals on the MB, the SSR light turns on and the bed begins to heat. I check the input + to – terminals and the reading is 24 V. So when the MOSFET fails, the SSR is energized. It should be just the opposite.

New Board Installed

    1. After the new board was installed, I turned on the printer and the SSR had no light on. SSR was not energized. I checked the input to the SSR and it was at 0 V.
    1. I set the bed temp to 30 C and the light comes on. I then checked the voltage between + and – output on the SSR and I get approx. 1.5 V. The input voltage to the SSR is 24 V and the bed is heating.
    1. Also, I checked the input terminals to the SSR at 0 C setting for the bed and I get 0 V. The SSR output terminals read 24 V. When the input terminals to the SSR are at 24 V, the output terminals of the SSR are at approx… 1.5 V.
    1. I am not sure how the bed is being heated, it seems to me that when the input is 24 V the output for the SSR should be 24 V to heat the bed. It is just the opposite.

I hope this makes sense.

I had been trying to use different coloured text to annotate your comments, but I couldn’t get it to work, soooo…

THIS IS WITH OLD BOARD

    1. The SSR is normally open by the schematic on the SSR. correct
    1. When I check the output of MB to the input the SSR, I get 24.22 V from my Fluke 77 multimeter. This indicates that the motherboard’s MOSFET is configured as a high-side driver and has internally shorted which is the cause of your problems.
    1. With the + on the SSR connected to V+ on the power supply and the – on the SSR going to the bed and I check from the + terminal on the SSR to the V- on the power supply, I get approx. 0 V. Am I reading this right? Are you referring to the SSR output terminals? With meter connected to the + SSR terminal which is connected directly to the Power Supply + and the the other end of the meter on Power Supply -, you get 0 Volts? Also, the terminals shown in the photo you supplied appear to be labeled L1, T1 (outputs) and A1 A2 (inputs). This is different from what you are describing. Can you post a picture of the new relay?
    1. With the old MB, when I turn on the printer, I get 24.22 V across the MOSFET on the MB. This is the input to the SSR. Therefore, this would enable the SSR. “Across” the MOSFET? Just to be absolutely clear, did you actually locate the MOSFET on the motherboard, identify it’s Drain and Source terminals, and take a measurement across them? I would think that should be 0V given that the MOSFET is behaving like it’s internally shorted; or by “across” did you mean you took your measurements across the motherboard’s output terminals that go to the relay? If so, it’s the same measurement you took in step 2. And, yes, that would enable the SSR.
    1. However, when I check across the output terminals of the SSR, I get 24.22 V, but the bed is not heating. That means the SSR is NOT being triggered and hence, the bed is not heating; but wasn’t that the original problem(?) that the relay was being triggered as soon as the power was being turned on? Or were you testing this with the relay disconnected from the motherboard? If you were doing this test with the motherboard disconnected, your readings are normal. Any type of switch/relay/etc. will always read the full applied voltage across it’s terminals when the terminals are open.
    1. If I connect the input terminals to the terminals on the MB, the SSR light turns on and the bed begins to heat. I check the input + to – terminals and the reading is 24 V. So when the MOSFET fails, the SSR is energized. It should be just the opposite. OK, so it sounds like Step 5 was without the motherboard connected and step 6 is with it connected. If so, everything is normal and indicates a shorted motherboard MOSFET.

New Board Installed

    1. After the new board was installed, I turned on the printer and the SSR had no light on. SSR was not energized. I checked the input to the SSR and it was at 0 V. Normal
    1. I set the bed temp to 30 C and the light comes on. I then checked the voltage between + and – output on the SSR and I get approx. 1.5 V. The input voltage to the SSR is 24 V and the bed is heating. Normal, although Rds(on) (Resistance between the Drain and the Source when the SSR’s internal MOSFET is ON) is usually in the milliohm range, so I’m surprised you’re getting a 1.5V drop across the terminals. I would have thought it would be significantly lower.
    1. Also, I checked the input terminals to the SSR at 0 C setting for the bed and I get 0 V. The SSR output terminals read 24 V. When the input terminals to the SSR are at 24 V, the output terminals of the SSR are at approx… 1.5 V. Normal. It means the SSR is not conducting when it’s input is at 0V and is conducting when it’s input 24V.
    1. I am not sure how the bed is being heated, it seems to me that when the input is 24 V the output for the SSR should be 24 V to heat the bed. It is just the opposite. That is actually normal. Think of the SSR as a mechanical, metal, switch. When that “switch” is closed, it’s shorting out the two output terminals, allowing current to pass from one terminal to the other. Of course, that also means that the voltage at each of those terminals would now be the same, since they are shorted together; hence, you are reading 0V (approximately) when the SSR is ON and the full applied voltage (24V) when the SSR is off (because no current is flowing and no voltage is being “dropped” across a load (the bed)). I think this is your main point of confusion.
1 Like

Obviously some of what I posted did not make sense.

I will post another picture of the SSR tomorrow. Also, I’ll read the current going through the wires to the bed.

When I am measuring the input to the SSR it is the same as across the terminals on the MOSFET on the board. I don’t understand why this voltage is 24 V when the MOSFET fails and not zero. Probably my lack of understanding electronics. The output of the SSR makes sense now. Open contacts of a electro-mechanical relay would be 24 V and closed should be 0.