Posts made by wylley

I'm looking to get some linear rails for a delta but I've read some conflicting things about the preload they should have.
I mostly see people saying it should have a light preload for precision but others say that it should have no preload for smooth movement on a belt. If someone could clear this up for me it would be much appreciated.

I'm still having trouble with the connector of the extension cable for my atx psu heating so I thought I could use the two different 12v rails and wire them like they were different power supplies for the bed and board. Turns out after mesuring with a multimeter that all the connections of the power supply are connected to a single rail and that the "two 12v rails" is only labeling/advertisement. Can I still wire it like it was 2 power supplies to split the current between the wires even tough its just a regular single rail power supply or not ?
Thanks for helping

@veti yes I have two 5V fans but even when I put them at 100%, it doesn't change anything, they only draw 0.05A each.

@veti using the 5v rail as external 5v the PSU fan doesn't stop spinning anymore and voltage stays around 11.6V whereas previoulsy it was around 11.2V . So I don't know how much the board draws but it seems that its more then the 0.9 A from my resistor. Maybe I can get stable 12v with an even bigger load, will try a halogen lightbulb.

The connector from the cpu extension cable I'm using is still getting warm though, even when only heating the bed to 60°c, so I still should put the bed on the 2nd 12v rail by connecting it to the board like its a 2nd psu as you did. Why did you need mosfets ?

@dc42
@veti
Currently im only using one of the two 12v rails, I guess I could take a pci-e cable and connect it like the wiki says to do when using 2 power supplies to put a load on the 2nd one. My only concern was the wiki saying to " not connect the bed heater PSU +ve output" and in a way it would be, but I guess thats just a technicality and it should be fine.

I will try using the 5V rail for external 5v and if its still not enough I will buy 25-50W halogen lightbulbs to draw more current.

@jamesm

• 120W heated bed
• 3x stepper motors rated at 2.5 A
• geared stepper motor rated at 1.3A
• 30W hotend heater
• 12V 0.6W fan

I'm trying to use a 730w bequiet L8 pure power atx power supply for the bed and hotend so I wired the 8 cpu cables to the board (4 x12v and 4 x GN) wich should give it 360w (it has two 12v rails rated at 30A each). When I heat even only the bed just to 60°c it draws so much current that the fan of the power supply stops and the cables get warm (would say ~40°c).

I tried putting a 5v load by putting a 5.5ohm 20w resistor on a molex connector and the fan does doesn't stop spining anymore but I got a low voltage warning (8.4v) and the drivers shut down because of overheating. The power supply is haswell compatible so I shouldn't even need it.

The only thing that worked was limiting the PWM of the hotend and bed to 60% but it's awfully slow and I got a heater fault once (hotend shut itself down mid print).

Thanks for helping.

@oliof you were right I forgot the dot: it's 39.8 and not 398 wich would be an insane offset. I can't beleive I copied this multiple times without noticing. When the error was talking about delta parameters it threw me off but I guess the probe offset is one of them.
Thanks a lot !

@oliof yes I followed it, the only thing I didn't do was mesure different trigger heights since the guide says that it is not necessary for first-time calibration and to return to it later.

here is latest bed.g that I tried (90mm probing radius) :0_1565405346043_bed.g

UPDATE: I tried entering the different trigger heights of the points in bed.g and it doesn't help, exact same behavior.
The problem is that the G30 commands like G30 X0 Y84.9 S-1 are not moving the nozzle.

Im trying to use autocalibration on a delta using a bltouch but I'im getting an unsual behavior : the nozzle goes down, then without moving probes the amount of points I had set and finishes by throwing "G32
Error: Auto calibration failed because probe point P0 was unreachable using the current delta parameters. Try a smaller probing radius."

I have tried down to a 50mm probing radius even tough my bed has a 125mm radius and double checked all the delta parameters but I still get no movement and the error. Looking at the coordinates of the probing points they should easily be reachable for the probe, the only thing Ive noticed is that if I put the nozzle at Z0 and go to Y25 for example the nozzle touches the bed (because its irregular/convex) but I dont see what that would have to do with the probing radius.

Thanks for helping

@veti After some research I've found that both of my thermistors had the wrong beta coefficient on Marlin since e3d says to use the TEMP_SENSOR 5 (wich has 4267 K beta value) and that my heatbed thermistor looks like a cheap chinese one with 3950 beta value. So I've put the e3d preset B4725 and C7.060000e-8 for the hotend and B3950 C0 for the heatbed on RRF.

I guess the manufacturer didnt care about the +-150 difference on both thermistors and just went with TEMP_SENSOR 1 on Marlin.

@veti It is a genuine one so I had the wrong Beta coefficient for years, good to know, thank you. I'm guessing leaving the bed at 4092 K like it was in marlin should be fine.

@Veti
Oh ok, this is really confusing so both of my thermistors were at 4092 K but the configuration tool sets it to 4725 when I select the e3 hotend preset which happens to be really close to the pullup. So what should I do ? Put both at 4092 like in Marlin or maybe only change the hotend's one to 4725 ?

@veti I just realized that in Marlin, I have both TEMP_SENSORs set to 1 and the comments say that "1 is 100k thermistor - best choice for EPCOS 100k (4.7k pullup)". So I assume that "pullup" is the Beta coefficient and so I will set the bed's one to 4725 too so its the same as the hotend's like in Marlin.

Makes me think one could do a simple script to migrate from Marlin to RRF.

Thank you all for the very helpful and quick replies !
The last concern I would have is the Beta values for the thermistors,
for the hotend I just selected the e3d hotend preset since I have an e3d v6 but I have no clue what the coefficient is for the bed thermistor and have no reference for the model, is it ok if I leave it as default (4138) ?

I'm trying to migrate from an old marlin firmware to reprap but I'm having trouble filling in some values since I cant seem to find them in the old marlin configuration.h

The first one being "maximum carriage travel", now I assume that I can mesure that by pushing one carriage to the minimum possible by going to an extremity of the bed, and it is roughly 440mm but I wonder if this is one of the values that the firmware uses to do precise math or just a safety limit that It wont allow the carriage to go further down. So, is there a formula to calculate a precise value or should I just put my rough mesure with a bit of wiggle room (like the printable radius) ?

-I can't find max speed (mm/s), if I understood correctly it is the same as the DEFAULT_MAX_FEEDRATE wich is just an old cnc machine term ?

-From what I've read the current limit of the A4988 drivers is vref*2.5, so I should use 1.25 correct me if I'm wrong.

-Lastly I would like to know if the huge diffence in the default acceleration ,RRF has 1000, my marlin 3000 and that, oddly my steps per unit are a bit higher (100 vs 80) but my extruder is significantly lower (450 vs 663) , is okay ?

I'm sorry if these questions have already been answered, I did my best to research them but only found vague answers.

I attached the Configuration.h (as txt) that I was using in marlin if that can help in any way.
0_1565152375158_Configuration.txt

Thanks a lot for helping,
Wylley