Hacking the HP45

Powder and inkjet printing
rofl
Posts: 56
Joined: Mon Aug 26, 2019 9:13 pm

Re: Hacking the HP45

Post by rofl »

dragonator,

I finally managed to run software on PC instead of RPI. I executed the same python file that was used on RPI, so no changes to software compared to both versions.

However, PC still produces vertical lines. Here is the log for one specific print and here is the image of this outcome. It is interesting that usually only the second swipe produces those vertical lines. I also tried other DPIs to print the same image, i.e. lower to 200, but those vertical lines were still present.

Do you have any ideas? It is weird as when I was printing last year, during some test prints, I did not notice these vertical lines at all... If those lines had been back then I would have definitely raised this issue... What are the chances that it has to do something with electrical part? As I checked voltage on head, it was around 11.8V, I decided to plug completely separate power supply to the head with 12.3V. I did one print with this higher voltage and those lines were still present. So not sure what can cause this....
Attachments
Screenshot 2023-01-04 201616.jpg
Screenshot 2023-01-04 201616.jpg (54.29 KiB) Viewed 691 times
Screenshot 2023-01-04 195907.jpg
Screenshot 2023-01-04 195907.jpg (29.09 KiB) Viewed 691 times
log_bj.zip
(2.69 KiB) Downloaded 44 times
User avatar
dragonator
Site Admin
Posts: 581
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

The weird errors are gone and this looks like a perfectly normal sweep to me. I have to say I am at a complete loss here. I have absolutely no clue why the controller prints these lines. They do seem to correspond with certain parts of the print, but this could not be the controller printing other lines in the wrong position, encoder error would just lead to messed up prints. If the offset is around 4mm it could be swapped primitives and addresses, but the distance looks wrong, and it seems to be complete lines. I am going to look in the code if I can find anything that could cause a full ejection on a side without any trigger.

For my reference, what software, controller and firmware version are you using, and did you make any modifications at all to anything. Did you try going back to a fresh download already?
rofl
Posts: 56
Joined: Mon Aug 26, 2019 9:13 pm

Re: Hacking the HP45

Post by rofl »

I really did not change anything. I double checked that and only things I changed were "setup" based, i.e. like build piston position, feeder piston position and speeds etc. No significant changes were made. Moreover, regarding teensy code, I barely changed anything as far as I remember.

I am using Oasis controller from the make package 20181230. Controller is V3.02 I think (this was the last version, wasn't it?). HP45 firmware, I think it came from the same make package and it should be V3.00. I checked that there is a newer make package, but quick review did not indicate that there some differences in firmware and software.

By the way, as far as I remember, inkjet commander still is not able to print, yes? I have new controller board working, but as far as I remember you said that you still did not manage to get it running with printing, yes?
User avatar
dragonator
Site Admin
Posts: 581
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

Commander prints fine, but it only sends bitmap images to a controller. It is not capable of controlling a machine to move the printhead. The new controller board runs fine on Oasis controller, it is just that the old controllers miss features used by inkjet commander.

New stuff is mostly backwards compatible, the old stuff is not forward compatible.
rofl
Posts: 56
Joined: Mon Aug 26, 2019 9:13 pm

Re: Hacking the HP45

Post by rofl »

So do I understand correctly, that I can run old Oasis control software with new controller board (including head movement)? Would it make sense to try in the first place?
User avatar
dragonator
Site Admin
Posts: 581
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

The issues do not seem to come from the software, and even though I have no clue why, they seem to be coming from the controller. If you have one you can try, but I would not make one just to test this.
rofl
Posts: 56
Joined: Mon Aug 26, 2019 9:13 pm

Re: Hacking the HP45

Post by rofl »

I have one V4.0 controller fully working and it can be just plugged on the printer. I will give it a try after I solve todays bunch of failures...
rofl
Posts: 56
Joined: Mon Aug 26, 2019 9:13 pm

Re: Hacking the HP45

Post by rofl »

dragonator,

I changed controller to V4 and it seems that initially vertical lines are not present anymore. However, conditions were not identical: some steppers were disconnected and not all devices attached (power conditions might not have been identical), so it might need some more testing, but so far, at minimal setup it seems to be OK.

However, I was not really able to setup V4 properly. Could you help me with that? As I understand, it needs some parameters to be defined, but I am not really sure what needs to be done in order to make it work in conditions that are identical to Oasis setup. When I connected V4 controller, it failed to set correct position. I got it that I need to setup correct source of motion, i.e. encoder. It started to calculate position, but it was in the wrong direction. I was not able to find, how to change that in an easy way, so I changed some settings in firmware:

#define POSITION_ENCODER_PIN1 34 //33
#define POSITION_ENCODER_PIN2 33 //34

Encoder positionEncoder(33, 34); //(34, 33); //make an encoder instance

I just swapped pins of encoder, which is still weird as I connected a new controller board in the exactly the same way as V3.02. When this change was made, position was correctly displayed and printing was possible, however, head missed some swipes and swipes were incorrect to be honest like first one was OK, second was printed in a wrong location and third one was like OK.

Is there a list of parameters than need to be defined for V4 controller? Can we find out what I need to setup to make it comfortably running on Oasis like setup with encoder and Oasis Controller software (not inkjet commander)?
User avatar
dragonator
Site Admin
Posts: 581
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

I was not aware of the swapped direction of the encoder connector. The fix you proposed will work fine, but it is not one I can push to the new firmware since it might break other setups. Another fix is swapping A an B channel, but all are a bit iffy.

incorrect swipes do sometimes happen on the first layer to me. It is an issue I have not been able to track down rightly. On later layers it should not happen, nor should swipes be missed entirely. If they are missed, that might indicate that the position the printhead moves to to start the sweep is close to the edge of the image. Does the printehead have distance to speed up, or does it start printing immediately after it starts moving?

I do not have a list of parameters at the moment. As the relative silence on my side might suggest, I am not spending a great deal of time on the HP45 controller right now. I am currently scaling the amount of time I am spending, and will spend back. There is a more recent version of the Oasis controller software aimed at controlling from a Pi screen. It never worked well on a Pi for me, but it has some quality of live features added to it. It is on the hackaday.io page, named compact: https://hackaday.io/project/86954/files

I do not know if it runs better, but for me it gives no real issues. It should also work on a V3 controller, the software was written when V4 was not around.
ghvanhoosesr
Posts: 6
Joined: Tue Nov 15, 2022 2:59 pm

Re: Hacking the HP45

Post by ghvanhoosesr »

Greetings Dragonator,

With the V3 printhead controller, is there anything that can be done for a problem where it looks like some nozzles are firing and some aren't?
I've gone through and verified the addresses and primitives are being sent on the software side. (the controls for the address latch are being sent by the Teensy when it should, and primitive pinouts on the Teensy are going high when they should) It almost looks like whole addresses aren't being enabled somewhere between the Teensy pins and the nozzles while others are working fine. The result is the printing comes out in thin horizontal stripes, each one with a millimeter or two of blank space between them.
I tried increasing the VHD voltage to 15V, still no effect. Also it seems to get warm, regardless of VHD voltage. Any thoughts you have on this issue would be highly appreciated!
Post Reply