Duet Wifi and CR-10

deckingman

This post is deleted!

deckingman

Above post deleted because I was stupid and got my maths wrong

If you were going to change the bed heater, it might be worth changing the printer to 24V. I use a mains heater but for a bed your size, a 24V heater would likely give you the heating power you need. It's a bit safer than mains power and saves the cost and complexity of a SSR. 24V @16 Amp is 384 Watts. You'd need to change the hot end heater and possible fans but there are ways to run 12V fans on a 24v system - it's just a jumper if all the fans are 12v.

opentoideas

@deckingman I understand the safety side of mains heaters and the benefit and simplicity of keeping it all 24V. some of the conversions I have seen online are just scary and while wiring is simple and will work safe is not a word that should be used for them! watched one yesterday with a chap using the low voltage CR-10 connector and not even considering that if unplugged the pins are live and easily touched! but it worked!!!

(just finished a 3 phase 24KW silicone pad heated plating tank running at 230C, man was that fun to ensure fail safe and safe operation! but agreed 24V would be the simplest and safest option for most)

To be fair have not yet decided how I will go about this however I have just dug out a 200W Meanwell PSU I had spare and will most likely use this though I do need a new hotend heater first ( I did read its possible to have 12V hotend with 24V system but its not ideal so I will wait for the cartridge before doing this.(its only a few pounds and the benefits seem well worth it)

given I already have this PSU and SSR's from other projects I am not sure that 24V bed heater is the best option for me but there is a lot to be said about keeping everything 24V and it should always be the first option.

deckingman

@opentoideas Another consideration is noise. You'll likely find as most people do that the machine runs very much quieter with the Duet. Then things like fans start to become intrusive and many of us have then started to use quiet fans but this is sometimes difficult to do with the PSU. So running a mains heater enables one to use a fanless PSU of around 200W or less. It sounds like you know what you are doing with a mains heater, in which case that might be a better choice. But there are other advantages to running 24V. I believe that steppers can be driven faster before the start to lose torque for example. I'm just trying to point things out but you'll have to weigh up the pros and cons and decide what is best for you.

HTH and best of luck.

opentoideas

thank you deckingman, your input is appreciated and it is reassuring to have others thinking along the same lines as I am.

the consensus seems to be that going to 24V is well worth it for many reasons and as I have the PSU I would be daft not to but as there are a few other bits that need to be changed to work on 24V this may have to wait for now but it is certainly on the list.

please let me know anything you can think of, its all good, and I want to work things through and while I am not the type of person to take a suggestion blindly and run with it at the same time I am fully aware that "you don't know what you don't know" so I would rather work through 1000 things I may have already considered as you never know when that one thing you didn't know will come along and bite you on the ass

opentoideas

waiting sucks - Duet on the Fedex van for delivery but not here yet lol

currently thinking over temperature monitoring....

so thermistors and PT1000 connect direct

thermocouples and PT100 require a daughter board to work

most of my printing is PETG so for the moment high temps should not be a problem and with only one machine consistency is more important than accuracy I presume. (again please remember I am newbie so feel free to correct me if i am talking out of my rear!)

assuming that despite accuracy being poor the thermistors are at least consistent then is there any reason I should swap as the cost is not insignificant. I am also assuming that I can add a correction factor to the temp reading in software if needed?

if so then any suggestions for ones that will work with the volcano's pocket?

deckingman

@opentoideas Based on my own experience, I bought both PT100 and thermocouple daughter boards. Never used thermocouples. Tried PT100s (not the E3D ones I hasten to add, as they wouldn't fit in my Diamond hot end). The ones I used had very thin wires which kept breaking and replacing them started to get too expensive. Changed to a simple thermistor and never looked back. OK so even if it's 5 degree out, it doesn't matter. When you try a new filament, you do some tests to determine the best temperature to print it at on your machine. So if you find 225 deg C to be best, that's what you use. It doesn't matter if the actual true temperature is 220 or 230.

I still have the second PT 100 connected to my bed but it picks up interference from the Z motor wiring somewhere that I've never been able to trace. It's no big deal - I just get short spikes to 2,000 deg C when moving te bed a long way such as starting a new print with the bed a long way down. I keep meaning to change it for a thermistor but simply haven't got around to it.

E3D thermistors would be the best compatibility wise. They'll fit and the data is readily available. But any known 100K NTC should do the job.

opentoideas

@deckingman - thanks again, this mirrors my own thoughts. I had a read the post where you also asked this and noticed in DC42's description of the types that though the PT100 is most accurate but that it could suffer interference.

shielded cables should deal with this but it's an added hassle.

I guess if you have more than one printer or farm out jobs then accuracy is most important as you want repeatability between machines but I was thinking along similar lines as you that as long as the temperature reading is consistent then the actual value is not so important for me as it would be for some.

I have some thermistors and PT1000 on the slow boat heading my way so it will be interesting to see if there is a difference.

I am more familiar with thermocouples for industrial use but they are a pain as you really need to have the same wire from the junction to the sensor so demounting the hotend is less simple.(though I have got away with connectors in less critical applications this defeats the point)

why is it that when I didn't need the parts the Fedex delivery was mid morning and now I need the part it's looking like I will get it at the last possible moment of the delivery day!

Murphy strikes again!!

opentoideas

next thing to think about.....

I currently have dual Z axis drive motors and the current wiring has these in parallel.

Looking at the Duet board it seems this has been considered and there are 2 Z axis headers and the advice I have found is that these are wired in series and this is better for our use.

so if I am understanding how stepper motors work then am I correct in thinking that if I use these 2 connectors one for each motor then the "A" coils on the two motors are linked in series and the "B" coils are linked in series?

if so then I just have to make sure that the wiring looms board to motor are the same for each motor to avoid them going in opposite directions?

is there anything I have not considered or am I just plain wrong?

deckingman

@opentoideas I'm not the best person to answer that as I only use a single Z stepper. What you say sounds right but to be sure, I think you'll find all the answers you need here https://duet3d.dozuki.com/Wiki/Choosing_and_connecting_stepper_motors

opentoideas

@deckingman lol not quite. been there and read that but it only got me as far as I am. hopefully there are others out there that will read this but either way I can test without the leadscrews connected just to be safe.

starting to look like Fedex re not bothering to deliver to me today - only 15 min left of the delivery window so not very happy with them. think I will be complaining to E3D in the morning - not that it's their fault but the couriers dont give a damn about the recipient only their client

dc42

@opentoideas said in Duet Wifi and CR-10:

next thing to think about.....

I currently have dual Z axis drive motors and the current wiring has these in parallel.

Looking at the Duet board it seems this has been considered and there are 2 Z axis headers and the advice I have found is that these are wired in series and this is better for our use.

so if I am understanding how stepper motors work then am I correct in thinking that if I use these 2 connectors one for each motor then the "A" coils on the two motors are linked in series and the "B" coils are linked in series?

Correct.

if so then I just have to make sure that the wiring looms board to motor are the same for each motor to avoid them going in opposite directions?

Yes.

However, if they are wired in parallel in the wiring loom, I suggest you leave them that way. If they are low current motors, then parallel wiring is better than series anyway.

Alternatively, if you have no other plans for the E1 motor output then you can connect one to Za and one to E1, then if you have a Z probe on your machine, you can get them to sync up automatically. See https://duet3d.dozuki.com/Wiki/Bed_levelling_using_multiple_independent_Z_motors. Use the M584 command in config.g to tell the firmware that you have them connected this way:

M584 Z3:5 E4

insertnamehere

@opentoideas said in Duet Wifi and CR-10:

Looking at the Duet board it seems this has been considered and there are 2 Z axis headers and the advice I have found is that these are wired in series and this is better for our use.

Another option is to wire each Z-axis stepper to its own driver so they can move independently. I have one Z stepper going to the Z driver and the other connected to the second extruder driver.

One of the problems I had with the CR-10 was that the X axis rail was difficult to keep level with the bed. If one side of the x axis was slightly higher or lower than the other it would affect the quality of the first layer. The level of the X axis would be different every time the printer was powered up and difficult to manually level.

I solved this problem by independently driving the Z axis steppers and I added some g-code to the slicer that would, at the start of each print, move the carriage left and right on the x axis, and using the z-probe, adjust each side of the x axis to be level with the bed.

Generally, I find this will get the x axis level to within +/- 0.003 mm and make a good first layer much more repeatable.

Oh, I see dc42 beat me to it.

opentoideas

@dc42 thank you and this is one of the many reasons I headed this way as every time there is something to think about it seems you have already thought about it figured it out and come up with a solution!

I now have the board and have to say if the initial network config is a taste of things to come then this might actually be fun.

if its not too much trouble would you be kind enough to check my logic on the CR-10 motors? the part number on the X and Y is JD42HS40-1004-02F though I cant find much I believe the 1004 refers to 1.0A so set up at 700mA

I know the E drive as its not standard and is the Titan aero pancake 17HS08-1004S so again 1.0A 1.8deg

but the Z motors just have Creality 42-34 which is not much help. Guessing they will be 1.8deg and possibly set a bit lower at 500mA?

opentoideas

@insertnamehere thank you so much, that sounds like a much better plan (even if DC42 was a bit quicker on the draw)

deckingman

@opentoideas said in Duet Wifi and CR-10:

I know the E drive as its not standard and is the Titan aero pancake 17HS08-1004S so again 1.0A 1.8deg

I could be wrong but I thought E3D's Titan pancake motors were 0.9 degree, not 1.8 - might be worth checking. It won't affect what current you set them to but it will affect the steps per mm (roughly 400 for a Titan with1.8 degree motors, but roughly 800 for the 0.9 degree motors).

opentoideas

@deckingman thats what I thought too but this may have been an oddball as it came from DjDemon and when I looked it up it came back with 1.8deg - https://www.oyostepper.com/images/upload/File/17HS08-1004S.pdf

that confused me when I first fitted it as I went with 0.9 rather than checking lol 837 steps vs 418 made a lot of mess!

dc42

@insertnamehere said in Duet Wifi and CR-10:

One of the problems I had with the CR-10 was that the X axis rail was difficult to keep level with the bed. If one side of the x axis was slightly higher or lower than the other it would affect the quality of the first layer. The level of the X axis would be different every time the printer was powered up and difficult to manually level.
I solved this problem by independently driving the Z axis steppers and I added some g-code to the slicer that would, at the start of each print, move the carriage left and right on the x axis, and using the z-probe, adjust each side of the x axis to be level with the bed.

Are you using M671 and G32, or something different?

opentoideas

so does this look correct in my config.h to set up the dual Z motors?

; Drives
M569 P0 S1 ; Drive 0 goes forwards
M569 P1 S1 ; Drive 1 goes forwards
M569 P2 S1 ; Drive 2 goes forwards
M569 P3 S1 ; Drive 3 goes forwards
M569 P4 S1 ; Drive 4 goes forwards
M584 X0 Y1 Z2:4 E3 ; Driver 0 controls the X motor, 1 controls Y, 2 and 4 control Z motors, 3 control E motor
M350 X16 Y16 Z16 E16 I1 ; Configure microstepping with interpolation
M92 X80 Y80 Z400 E420 ; Set steps per mm
M566 X900 Y900 Z12 E120 ; Set maximum instantaneous speed changes (mm/min)
M203 X30000 Y30000 Z900 E1500 ; Set maximum speeds (mm/min)
M201 X500 Y500 Z100 E5000 ; Set accelerations (mm/s^2)
M906 X800 Y800 Z500 E800 I30 ; Set motor currents (mA) and motor idle factor in per cent
M84 S30 ; Set idle timeout

I have wired in and tested the heaters, thermistors, end stops and fans and am about to start on the motors. I am sure I will have to correct direction on some but I plan to do that in the testing phase.

I am really glad there are plenty of crimp terminals as I had a few losses until I got things dialed in but I have to say the setup and testing so far has been a dream

dc42

Yes that looks good to me (I don't know whether the steps/mm are correct for the CR10). Eventually you will want to add a M671 command as well, but that can wait. The M203 values may be too high, again I don't know the CR10.