headbed with glass plate lag ?

Hi all

I just got my heatbed and I installed it. I am measuring the temperatures on top of the glass plate with my multimeter, and the temperatures
are correct but it has a descent amount of lag a minute per 10c .
Is there a way to suppress this. or should I just program a wait ?

That's what happens when you put a thermal insulator like glass on the bed. You wait.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I think that's real, i.e. the top of your bed is actually a different temperature from what the sensor is picking up. So, just wait, if you want the temperature to be equal.

Ok, thanks. I will see how I need to do that with Repetier

So I need to add this to 'Start gcode'

M140 S115 ; Set bed temp, no wait.
M104 S225 ; Set nozzle temp, no wait.
M116 ; Wait until target temperatures for extruder and heated bed are reached
G4 P12000 ; wait

@the_digital_dentist, I know: but before I saw your new design I thought this was the best solution, to avoid thermal expansion, and I ordered everything, So I need some prove that I was right .
If this works I may start looking at new design.

Edited 1 time(s). Last edit at 03/11/2016 02:31PM by amigob.

mhhh, that didn't work, it didn't wait for a second ??

Try:

M190 S115
M109 S225

The first command sets the bed temp and waits until it reaches 115C before going to the next command, then the extruder will heat up and by the time that is done the top surface of the bed should have reached print temperature.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
amigob
mhhh, that didn't work, it didn't wait for a second ??

Sounds like a firmware bug to me. What firmware are you using?

---

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].

Repetier I tried first ( when it did'nt work via the Repetier start code.
Maybe I need to activate it some where ?

If I add it to the Slic3r Start code it will work. But I changed a little bit

M116 ; Wait until target temperatures for extruder and heated bed are reached
G4 P120000 ; wait

the temperature settings overrides the one from the Slicer produced gcode So that is bad :-)
And I added a '0' 12 seconds was not enough wait :-)

dentist, I will try that also, maybe that works better.
Starting to get bed time here :-)

Thanks

I found another solution for this lag problem,
I am going to mount a contact less temperature sensor underneath my x carrier.
This sensor will give me the feedback of the real temperature of the bed. ( tested it with a Velleman contact less temperature sensor ). and it can also provide the uniformity of the temperature of the bed.
Ok, I need to add some code to the firmware but that is no problem for me.

You can even measure the temperature of your object with it, maybe also useful :-)

So ordered the sensor.

MLX90614BAA

It only needs 3.3 volt and I2C, and then you can read the temperature.
So I only need to add 2 wires the 3.3v was already available for the bed height sensor.

How does having the "real" temperature of the bed, as reported by an IR sensor, reduce the lag in heating the top of the bed when you put a piece of glass on it?

Why doesn't the thermistor that you're already using tell you the "real" temperature of the bed?

IR temperature sensors are only accurate when pointed at an object of a specific reflectivity. A glass plate is very reflective and will probably cause the thing to give an incorrect reading. Using the top of the glass temperature to set the bed temperature will probably result in a pretty big overshoot of the target temperature due to the lag that the glass plate introduces.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Cool toy though! I like the idea of being able to walk it across the bed and build up an IR image. Not sure it's much practical use but it sure would be interesting to play with.

I doesn't reduce the lag of the heat transfer but you can start printing at the moment that the bed is at the designate temperature. So I don't have to add a delay any more, it just waits until the Glass plate reaches the designated temperature.

I was also surprised that the hand contact less temperature sensor sensor was giving a reasonable correct temperature, compared it with my multi meter with a thermocouple( even when tilting the sensor it still produced the same value ). I have painted the back of the glass plate black, maybe that helps a lot.

Test show that it works a lot better then the sensor in the heater. But the sensor in the heat will still be used in the control loop when printing.

So I will adapt Repetier to use this sensor in a preheat scenario, so that the printing starts when the top of the bed reaches the designated temperature.

Let us see how it will work, I will definitely help me, because I have now a delay of 5 minutes. after the heat bed sensor reaches 110c( that is after 1.5 minutes),

Thinking now about it, and I can reduce the lag time, by increasing the temperature above designated temperature until IR sensor reaches the designated temperature, and switch the heat bed back to designated temperature when reached.

You may be over-thinking it a bit. It's not just that there's a delay in the temperatures between the top and bottom of the bed, unless you insulate the top surface (which we can't if we're going to print on it) there will always be a temperature gradient from top to bottom. It doesn't matter, as the target temperature is just a nominal number that people have arrived at by trial and error using systems with very similar temperature gradients, and something you tune as you go along. The PID algorithms work best with the sensor close to the heat source, so it's best to keep the thermistor in the normal place, otherwise it would make sense to swap to a top mounted IR sensor like the one you suggest. It's interesting to know roughly what the difference between the set and actual surface temperature is under different conditions, but it's mostly an intellectual rather than practical exercise.

The time the printer needs to wait, is different every time, ( switch bed temperature, different start temperature of the bed).
Now the printer has a worst case delay of 5min. With this sensor The printer will wait until the glass plate reaches the designated temperature.

I will give you some examples :

If I set 110c, when the heat bed heater sensor reaches the 110c, the glass plate is ~85c. starting to print with ABS will not go correctly
because the ABS will not adhere to the bed that well. So with the 5min delay the glass plate will reach the 110c.

When I only need to reach 90c ( PETG ) I only need 1.5min delay to really reach the 90c on the glass plate.

When I start when the plate is 70c already then again I have to wait a lot less then the 5min to get to the designated temperature. This simple IR sensor will fix this, I don't need a delay any more because I will measure to top of the bed.

A second thing ( first use the sensor to remove the delay) is that I can increase the temperature of the heater to reduce the time until the glass plate is at the correct temperature, because now the heater will not go above the designated temperature. Speed up the process. I don't like to wait :-) because at this point in time I want to see the first layer been printed.

The printer is in the attic and I don't like to be there if not needed.

This is a bit of a strange definition "The PID algorithms work best with the sensor close to the heat source"
I don't care about how the PID or BANG BANG does its work. The goal is that the top of the heat bed needs to reach a specified temperature, reproducible, and with the sensor in the heater that is not possible.

Edited 1 time(s). Last edit at 03/31/2016 08:18AM by amigob.

If the worst case delay is 5 minutes, why not just wait 5 minutes before you start printing? Is it really worth the effort and added complication to add an additional temperature sensor and modify the firmware?

If you just used a flat aluminum plate to start with, you wouldn't need the glass and wouldn't have any delay to contend with.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I am really starting to like the glass/3DLAC spray combination. The prints stick extremely until cool down and they pop of without any effort or tooling. The only thing was the cleaning, but normal soapy water or glass cleaner ( I tried this because I think the spray is just normal hairspray ) works perfect.
So I will add this sensor, and it is compared to all the other software and electronics in the 3D printer really simple.