Help with temperature sensors on bUltimaker

[3dprntz 4]

I had the problem and the printer isn't working right now. I tried to update the firmware through Arduino with this firmware (https://www.thingiverse.com/thing:4223870) and everything worked until I heated up the bed. When I turn on the power the printer wont turn on, but everything turns on when connected to the usb cable. I had Marlin V2 already on it before but had the stock settings and thought I would add a few other updates. I then tried to reinstall the old hex file but Cura says it cant connect to the printer with the usb. Tried burning a bootloader, which I think is the thing that helps cura update the firmware to the arduino, but that doesn't work. Now I'm trying to install the github old verison of ultimaker marlin on the printer. I've uploaded it but I still can't update through cura nor does the file compile when I try to change it to reprap full graphics display. I have all the libraries already installed and the original version is good at already declaring that, but it says that the directory for utility/U8g.h doesn't exist. The other thing is I don't think that Im installing the bootloader correctly, or maybe I am?

 

If it helps, Ill give specs on my printer:

-Power supply: 19V (stock)

-Extruder: Stock

-Display: Full graphic Reprap display

-Heated Bed: 18~24V, 200 W, With thermistor (still unsure what the thermistor is, but I put it as 100K thermistor or option 1)

-Arduino mega 2560 board (stock)

[gr5 2,168]

I strongly recommend you don't mess with the bootloader.  It may seem like it's gone but this is a common thing with arduinos - it seems like the bootloader is gone but it isn't.  I have spent many frustrating hours with arduinos rebooting 1 second after starting compile, rebooting arduino 2 seconds after starting compile, etc.  sometimes that trick works.  Sometimes it doesn't.  If it *looks* like it loaded a new firmware then it did.  Either it works 100% or it works 0%.  There is no in between where the progress bar gets to 100% but it didn't work.  If the progress gets to 100% then it loaded properly.

 

the arduino IDE works better at loading a new firmware than the ultimaker GUI.  The Ultimaker GUI doesn't seem to like certain COM numbers (e.g. COM3 through COM10 seems to work but sometimes my arduino is at COM13).

[3dprntz 4]

Still trying to get the firmware uploaded. Im hearing that the Arduino ide sometimes has problems with the Arduinos due to the braudrate connection and some other things. Theres the timeout or authorization denied. Ill look more into it.

 

On 11/24/2020 at 6:00 AM, gr5 said:

Anyway to answer your question - just hook it up - you'll be fine.  If the power supply can't deliver enough power it will just shut down when you turn on the heated bed.  Nothing will get damaged.  It just won't work.

On another note, I believe the part when the printer shutoff from heating the bed was maybe what you said before. The bed with 19v power supply would have 1.805 ohms with 200W and 10.52 A. Would regular marlin have the option with the budget hotbed or is that only with tinkermarlin? Also, isn't tinkermarlin only for upgraded boards like UM2 and UMO+?

[gr5 2,168]

baudrate?  If you mess with the baudrate serial rate (change it from the default (115k?)) then I assume you have to load your project in that brief time before the baudrate is switched which may be within a second after power up.  So don't mess with the default baudrate if possible.

 

Arduino can be such a pain to program.  I've wasted many hours with this issue on many separate occasions.  And then other days I can load firmware 100 times in a row no problem (I've done a lot of Arduino development).

[gr5 2,168]

Just now, gr5 said:

So don't mess with the default baudrate if possible.

 

Let me clarify.  The usb/serial connection on any arduino can be used to upload new firmware but it ALSO has other uses such as debugging output.  If you change the baudrate in the code that is loaded onto the arduino (for example Marlin allows you to send gcodes over the same cable and the baudrate is set in the marlin firmware).  That's a bad idea (changing the baudrate in Marlin - it's probably in Configuartion.h or Config_adv.h or something like that).

[3dprntz 4]

I get it. Wish Arduino was a bit easier to connect and upload. I have the builds ready to upload but it times-out to connect to the Arduino, even though the IDE can get the Arduino info.

[3dprntz 4]

I got it uploaded. The one thing that happens it when the heated bed is turned on the printer turns off. It looks like its the power supply brick that turns off. Should I get a new power supply brick that is 19V? Or should I go back to the relay?

[Torgeir 242]

Hi @3dprntz,

 

If it shut down like this, the problem is in your power supply, -or in the wiring toward the printer - but most probably in the power supply unit itself.

 

Sorry but this below is only for UM2, Not UMO with 1.5.7 board...  To much UM2+ thinking..🤔

 

-----

 

If it do not shut down, but the K1 relay is faulty; -your heat bed, extruder heater and all of your stepper motors will neither work.

There is no warning that is reporting that the K1 relay is faulty!

However, as soon your try to increase the bed temperature there will be an error displayed as; "Error Stopped" - "Heater Error" etc., depending of what you're trying to do.. After this error you cannot do anything, but recycling the power switch will get you back to previous stage.

All things look normal until you try to print or making some test function related to the above functions.

 

-----

UMO.

You are talking about making the MOS FET transistor for the heat bed to driving the relay directly?

This do not change anything and your power reset will still occur..

You need a power supply that can handle this load..

 

Good luck

 

Thanks

Torgeir

Edited December 5, 2020 by Torgeir
Wrong pcb.

[gr5 2,168]

Quote

Heated Bed: 18~24V, 200 W, With thermisto

Yeah this is too vague.  How many ohms is it exactly?  I'm going to guess it's 200W at 18V so that means 355Watts.  it would help if you ohmed it out or if there was a more specific specification.  The wattage will change with the voltage so the question is "what's the wattage at 24V?".

 

So do you have tinkerMarlin firmware?  I assume not - that has some really nice features around power management.  It would easily allow you to never use more than 150W (default UM2 heater).

 

Please give the model number from your power supply.  I assume it's one of these two:

GST220A24-R7B
GS220A24-R7B
 

So if you want to roll your own Marlin and don't want to use a tinker version then you have an option in Configuration.h:

#define MAX_BED_POWER 255

 

255 means 100%.  You want to restrict to 150W and I assume it's 355W so you want to restrict it to 42% (150/355=42%).  42% of 255 is 107.  So set MAX_BED_POWER to 107.  This should hopefully restrict power to 150W and your brick should be hopefully happy.  It might not because 355W is going to draw a LOT of current when it's on.  Let me back up - setting it to 107 will bean it will turn the bed on and off about 20X per second.  So 21ms on and 39ms off.   Your lights are going to flicker like crazy when the bed is on but I think it may work.  But the printer will be pulling a LOT of current/power for 21ms before giving the supply a break for 39ms to recover.  That might be too much.  It will probably be fine.  The main capacitors (they are like batteries that store up the power during that 39ms recovery) may get too hot but I think they will be fine.  I'm pretty sure the supply can handle this.  The voltage will be dropping 20X per second - down to maybe 22V or something but hopefully/probably nothing will care.  The steppers won't care.  The heater won't care.  It  should work.

 

So:

1) What is the actual wattage of your bed at 24V?  (if you know the resistance, the formula is  wattage = voltage^2/resistance)

2) What firmware did you end up with?  If you have tinkerMarlin you can fix your problem in the power budget settings

3) Which power supply do you have?

 

There is one more option - get a bigger supply like you suggested.  I doubt you can find a 500W 24V supply for less than the cost of your entire printer (You need 355W for the bed, 25W for your nozzle (what is the nozzle wattage?) and about 25W for the servos).

 

And a final option is to get a supply just for the bed and add a relay and have that supply only power the bed.  If you get a 18V supply then the bed will probably be only 200W which is more reasonable.

 

Getting an 18V supply is not a good option as the servos work better at 24V but you can go that route also.  Still 200W is too much for most power supplies.  But a 300W 18V supply might work - but now you need a more powerful nozzle heater.

 

 

[3dprntz 4]

On 12/6/2020 at 9:33 AM, gr5 said:

Yeah this is too vague.  How many ohms is it exactly?  I'm going to guess it's 200W at 18V so that means 355Watts.  it would help if you ohmed it out or if there was a more specific specification.  The wattage will change with the voltage so the question is "what's the wattage at 24V?".

 

So do you have tinkerMarlin firmware?  I assume not - that has some really nice features around power management.  It would easily allow you to never use more than 150W (default UM2 heater).

 

Please give the model number from your power supply.  I assume it's one of these two:

GST220A24-R7B
GS220A24-R7B
 

So if you want to roll your own Marlin and don't want to use a tinker version then you have an option in Configuration.h:

#define MAX_BED_POWER 255

 

255 means 100%.  You want to restrict to 150W and I assume it's 355W so you want to restrict it to 42% (150/355=42%).  42% of 255 is 107.  So set MAX_BED_POWER to 107.  This should hopefully restrict power to 150W and your brick should be hopefully happy.  It might not because 355W is going to draw a LOT of current when it's on.  Let me back up - setting it to 107 will bean it will turn the bed on and off about 20X per second.  So 21ms on and 39ms off.   Your lights are going to flicker like crazy when the bed is on but I think it may work.  But the printer will be pulling a LOT of current/power for 21ms before giving the supply a break for 39ms to recover.  That might be too much.  It will probably be fine.  The main capacitors (they are like batteries that store up the power during that 39ms recovery) may get too hot but I think they will be fine.  I'm pretty sure the supply can handle this.  The voltage will be dropping 20X per second - down to maybe 22V or something but hopefully/probably nothing will care.  The steppers won't care.  The heater won't care.  It  should work.

 

So:

1) What is the actual wattage of your bed at 24V?  (if you know the resistance, the formula is  wattage = voltage^2/resistance)

2) What firmware did you end up with?  If you have tinkerMarlin you can fix your problem in the power budget settings

3) Which power supply do you have?

 

There is one more option - get a bigger supply like you suggested.  I doubt you can find a 500W 24V supply for less than the cost of your entire printer (You need 355W for the bed, 25W for your nozzle (what is the nozzle wattage?) and about 25W for the servos).

 

And a final option is to get a supply just for the bed and add a relay and have that supply only power the bed.  If you get a 18V supply then the bed will probably be only 200W which is more reasonable.

 

Getting an 18V supply is not a good option as the servos work better at 24V but you can go that route also.  Still 200W is too much for most power supplies.  But a 300W 18V supply might work - but now you need a more powerful nozzle heater.

 

 

 

I haven't gotten a great reading but the ohms on the bed have a range of .8 to 1.2, so I'm thinking its around 1. I'm doing some tests on it to see the actual specs as the site doesn't provide much information on the heated bed. My guess is that is is like the standard reprap heated bed that you find everywhere. Im checking also the possibility of changing the power supply to 24V and just using one means well PSU. Ive checked the power supply yesterday and its a PA-1121-04 Liteon power supply, which probably isn't the right one. So maybe the real power supply has been mixed up with another over the years from moving. But I've ordered a 350W 24V 14.6A power supply from Amazon to test the bed. It might work as a replacement for the actual one, but we will see. I also have a mosfet for the board right now, but maybe I can get it connected directly to the board when I convert it over.

The software I am using is the regular bultimaker firmware generator. Its been easy to install and work. I wanted to upgrade some of the systems in the Ultimaker in the new firmware, but I've had problems with the code and I haven't had time to look into it.

Edited February 19, 2021 by 3dprntz

[3dprntz 4]

I got a separate PSU and Its working great. its a 24V 350W. So heats up the bed quite quickly. I tried tuning the PID on the firmware, but don't think its being saved. Does the bUlitmaker Firmware support saving auto PID tuning on the heated bed? If not, does that mean I will have to create my own configuration to do it. Or is there a way to edit this in the file that you get from the firmware builder site?

[gr5 2,168]

Normally you don't need PID tuning for heated bed.  It's not normally going to oscillate more than 1 or 2 degrees even in "bang bang" mode.  Bang bang mode means PID is disabled and it's like a house thermometer - if it's even slightly above goal temp it's off.  If it's even slightly below goal temp it's on.

 

However you have a print bed that's around 400 to 600 watts (at 24V).  That's one hell of a bed heater!  Even if it's 1.2 ohms, at 24V that's 480 watts!!  You can easily get that bed to 180C (360F).  And hotter.  That's hot.  That's pizza oven hot.  That's hot enough to melt everything in your printer that's not metal.  So be careful!

 

If it's possible to use the bed with a 12V supply instead then that might be safer.  I'm almost certain that heated bed was meant to be used with 12V supply only.  But it will work.

 

There's also an option I think to reduce max power - certainly to the nozzle heater - but maybe also to the bed?  If so you can set the max power to 50% or even lower.

[3dprntz 4]

Guess I got to start making pizzas on my machine. hahah. Im not worried about the bed as much anymore. I think I might upgrade the machine to a 24V machine.

[gr5 2,168]

I thought it already was a 24V machine?  What is the voltage?  The wattage of the heater is V*V/R   where V is the voltage and R is the resistance.

[3dprntz 4]

the machine itself is 19V with the power supply and the amps is 6.32A, so around 120 Watts. When we did tests on the heated bed, we picked up around .9 to 1.2 ohms on the bed with the multimeter and couldn't get it consistent. Right now the bed is the only one connected to the 24V 350W power supply.

[3dprntz 4]

On 11/20/2020 at 1:58 PM, Juliogdiana said:

Got the printer fully working now with ultipanel controller, hot end temperature works perfect with amplifier e3d pt100 board, resistor R23 cut and firmware with sensor 20, same as um2.

 

I would like to have the full graphics if I can sort the firmware.

Looking to do this on my board for both the first and second temp sensor. How did you get the connection to work? Did you have to use a different ground, or once the resistor was cut, the direct connection worked for the temp to the e3d amplifier board?

[Torgeir 242]

Hi @3dprntz

 

All professional made PT100 connection wires are normally shielded with a copper braid. This copper braid is not connected to ANY of the two wires (or more wires) connected to the "precision platinum resistor".

 

The amplifier used for this sensor is of differential type, so no ground to any of those two inputs!

 

If there is a copper braid shielding, this wire is connected to ground to prevent unwanted electromagnetic disturbances to enter and give error to our temperature reading.

 

Thanks

 

Torgeir

[3dprntz 4]

If I am understanding you correctly, you didn't connect the ground on the  Temp 1 of the 1.5.7 board to the ground of the amplifier board and left it unconnected since the PT100 has the copper braid for ground? I'm using the same pt100 thermistor as the UM2 but testing it on temp 2.

[Torgeir 242]

Well, -if you use the PT100 directly on the Temp2 input of the 1.5.7 board, your indicated temperature will be way of the "correct value"..

 

If you plan to use the 3 temp input on this board using PT100, you'll  need to remove all 3 pull up resistors that

 are connected to each input line and you must use an E3D PT100 amplifier for each of the 3 inputs.

This is the three 4.7 Kohm (4700 ohm) resistors to be removed, R23, R21 and R4.

Then you can connect the E3D amplifiers output to each of Temp inputs on the 1.5.7 board.

 

Thanks

Torgeir