Temp drops when printing starts

[Pobo]

I just finished the build on my V2, with the standard Bowden thermistor. I've printed and installed the fan shrouds, reset the PID. Everything has been working great. I've done 5 great prints, plus the two shrouds. Now when I start a print, the nozzle heats up and holds the temp fine while the bed is heating. Once the print starts, the nozzle temp drops out. Matter control shows the temp target at 228 (ABS) but the temp slowly drops to room temp.
All connections are solid as far as I can tell. Do I possibly have a resistor in the thermistor out? If so, would you recommend replacing the resistor, the entire thermistor, or upgrading to the newer thermistor series?

[blackrabt]

I am also having this issue with the temperature dropping once printing starts. I just replaced my thermistor and the problem has not been resolved. What other causes should I look at? Even if I crank the set temp up once the drop starts it has a hell of a time recovering and if the fans are on it is a lost cause.

[IMBoring25]

First, check the duty cycle history for the hot end. If it is running less than 100% duty cycle when this is going on, the PID needs further tuning. Autotune is not an exact science.

If it IS running 100% duty cycle and still not maintaining temperature, a few things can contribute:

One or both fans blowing on the heater block can contribute. Insulating the heater block (Kapton tape isn't extremely effective, but is available after the build and can handle the temperatures) can help.

It's quite possible the power supply isn't delivering a full 12V when all the load is on it, especially if it's the old-style computer (ATX) power supply. Even with the bed running on its own power supply through a Solid State Relay, my main power supply is still delivering in the 11.2V range while printing. I'm having to use the LCD to slow down the print to keep the temperature high enough to keep filament flowing until I can get the power supply replaced.

You can disconnect the heater circuit from the board and check its resistance. The original V2 hot end uses two 6.8-ohm resistors in parallel, so the circuit should have a resistance somewhere near 3.4 ohms. If it's significantly different from this, something may be wrong with the resistors.

[DeltaCon]

If it really drops to roomtemp, it is not a matter of insufficient PSU. It is possible that a PSU can't keep up with the set temp, but dropping to roomtemp is something different, especially when the bed temp is holding up. The bed takes a lot more power than the hotend. Pobo are you sure the set temp still is 228 when this happenes?

[IMBoring25]

Good call. Read right past that detail. Wonder if it's timing out.

[Pobo]

Mattercontrol shows the current temp settings, which remain at 228 when the hot end temp drops. I re-tuned the PID after installing the fans. Being that I'm using ABS the print fan is off, which I've verified. As I said, the temp holds while the bed heats, then drops when the printing starts. The bed temp doesn't drop at all. I was also able to make several prints before this started happening.
I'll check the power supply output this evening, as well as the duty cycle history, and the Hotend resistance. I'm suspecting it's a resistor. If that's the case I'll probably order a couple spares.
I've been very happy with the quality of the prints I've done so far, and would like to maintain that level of performance. Is the new HE280 Hotend difficult to upgrade to? I'd like to get this one running again before thinking about upgrading anything, but it may be a future project.

[rurwin]

I'm an extreme newbie here, so I might be missing something, but I'm an embedded software engineer by trade.

If it's dropping to room temperature as soon as printing starts, but it works again immediately afterwards to preheat, then it almost certainly isn't a resistor and it definitely isn't a thermister. To check the faint possibility of the resistor, heat it up under manual control and run some filament through it. Jiggle the head around a bit with the manual controls. If it keeps temperature then no is cut-out is happening and there's probably no broken wires that come mysteriously loose at operating temperatures.

It would be highly unusual for the PID to work to heat to preheat and then fail to zero for working temperature. There is one PID algorithm I have used before that can produce that behaviour, although I wouldn't expect it to be so repeatable. If the PID is calculated in "velocity mode", basically as a differential of the calculation, then it requires some Integral value to stabilise the set-point. If that was your problem it probably wouldn't hold any constant temperature but the maths is screwy; ensure I is not zero. (But not too much or you'll get wild oscillations.)

Personal note: These PIDs seem loose to me. I'm used to controlling engine coolant temperature to 97C, when 100C means someone has to hit the emergency stop and get out the mop and bucket (they are run unpressurised.) And that with a varying heat input of up to a quarter of a megawatt.

[Pobo]

So I did what rurwin suggested. I brought it up to temp manually, then extruded and moved the head manually. It seemed to work fine until I lowered the head almost to the bed. That's when the temp started dropping. It didn't drop completely, but wouldn't hold extrusion temps. I'm going to assume I have a wire that's broken. I'll be changing it out soon. Won't have time until next week

[rurwin]

Check your wiring harness; that's what is stressed when the Z height changes. It might be as simple as unplugging all the connectors and plugging them in again or one of them might not be fully inserted.

[Pobo]

Problem solved. The lead wires to the hot end were loose inside the connector to the resistors. I pinched them tighter and they're holding heat well. If I have more trouble, I'll probably solder them and add connectors to the wires in case I need to remove the head.
Thank you all for the help and advice. All knowledge and ideas help and give me things to watch for.

[geneb]

Don't solder them. The heat will eventually cause the solder to evaporate out and you'll be back to an intermittent connection.

g.