Printable View
PICStep now has a V2.0 firmware available.
It has a new 1/16 mode (1/16, 1/8, 1/4 and 1/2, lost full), a 5 1/2 minute auto shutdown and the direction line bug fix.
The motor should sound exactly the same in either direction.
http://www.fromorbit.com/projects/picstep/
Alan.
Thank you.
Zoltan
Where is the v.2 software listed on the site? all i can see is v1.1
i think something happened to his hosting.Quote:
PICStep gets 1/16 stepping and a bug fix! 25/07/06
After deciphering what RadekCX from the Czech Republic was on about in numerous hard to understand posts on CNCZone, I finally figure out what he was talking about. He had found a bug in the firmware that was caused by a bug in the LMD18245 datasheet.
The LMD will not switch correctly if the direction of the energy flowing through the coil changes before the LMD DAC is told to go to 0 (it must be 0 before the direction change). A tweak of the firmware and some tidy-ups and this issue is now resolved. The motor should sound exactly the same in both directions of rotation (and may have less resonance at high speed).
I've also added 1/16 stepping because I was rewritting the DAC tables anyhow and it seems like a nice little upgrade. The mode switches now select from 1/2, 1/4, 1/8 and 1/16. So if you liked the 1/8 mode you will need to adjust you mode settings to keep it as it will now be 1/16 otherwise.
Also I've made two versions for the compiled HEX files. One with the auto 5 1/2 minute shutdown and one without for those that hated the shutdown feature.
Let me know if you find any more bugs!
* NEW Firmware V2.0 20MHz - Sources + HEX
* NEW Firmware V2.0 (No auto timeout) 20MHz - Sources + HEX
Cool thanks,
What setting is torque compensation on? 1/2 step?
A while back I ran into an issue with the 16F628A not programing properly.
I don't remember the issue. We were using firmware designed for the 16F627 as I recall and there was something different about the 16F628A that would not work.
Does this code work in the current 16F628A? Microchip is only offering samples on the A versions of the pdip package.
I'm currently running the 2.0 firmware on my 16F628A didn't have any programming issues, using mplab icd2
I do not seem to be able to get it to change micro step mode???
The jumpers hace no effect???
In the source there is only one dac table for 1/16 mode
I really need this thing to 1/4 step.
Did you try manualu, step-by-step? If using GCODE speed can be to fast to see difference, but with jog difference is so obviously.Quote:
Originally Posted by tfmacz
If the drive is in the correct mode a move of one inch should result in the same number of revolutions as the pitch of the lead screw and the carradge should move one inch.
on this machine a 1 inch move should result in 10 revolutins of the lead screw. I was only getting 2 1/2 revs.
As it turns out I was trying the "Non Auto Timeout" version. I tried the "Auto Timeout" version and it works properly, so possibly the "Non Auto Timeout" version is broken.
Thanks for your suggeston though.
Ted
Yes, I just tried with "Non Auto Timeout" and it didn't work. I.e. works, but has only one mode - 1/16 and jumpers has no effects.Quote:
Originally Posted by tfmacz
ASM codes are the same for both SW, except line 23 #define TIMEOUT_ENABLED and it is normal.
In any case, I like "Auto Timeoutcode", but I will be really happy if Alan will be so kindly to tell me how to change timeout from 5.1/2 minutes. I didn't know assembler well so I can't find "magic value" in code.
I'll have a look at that, the defines are probably hiding something they're not suppose too. Since I never use the non-timeout version I don't really test it to fullyQuote:
Originally Posted by kodalo
It's done using the 16 bit TMR and a few overflow registers. There isn't really a magic number to keep the code fast.Quote:
Originally Posted by kodalo
Comment out these lines to reduce the timeout to like 70secs or change "timeout_reg, 2" to some other bit field number (ie. 0-7) to increase the time. Experimentation is the key (else crack out the calculator and use the number of instructions versus the clock frequency to get an exact time)Code:incf timeout_reg, f ; Increase the timeout reg value
btfss timeout_reg, 2 ; Check we've been around the 4 times of the 16 bit counter (~5 minutes 45 seconds @ 20MHz)
goto Loop
-A.
Thanks..I would prefer to use the drive with out the timeout. There is a disable pin on the servo drives this will be working with that is controlled by a switch on the control panel and by EMC2 which is the controller software I am using. The "Enable(Brake)" pin will be used to "sleep" the entire drive.
The servo drives run the X and Y axis and this stepper drive will run the Z.
Thanks, but now I have problem with MPLAB. There is to many errors when I trying to compile. Is it assembler code correct?Quote:
Originally Posted by Garfield2
Seems the no timeout version still haven't fixed yet. Does anyone can modify the timeout (lets say 30mins) with sucess?
I have been using the "stock timeout version" here for some time now and it has worked perfectly and the "timeout" has not been a problem. It all depends on if there is enough "gear reduction" in the axis drive to resist moving with the motors turned off. For me the 3:1 reduction and the magnetic detent in the steppers is enough to lock the axis in place.
Quote:
Originally Posted by bearbears
Because my milling machine is using direct motor coupling and grounded ballscrews, when the motor sleep those axis will move by the cutting force. As I would like to use the 1/8 microstepping so that the no timeout version can not be used. Anyway, thanks for your tips and reply.
Ok guys i have been using picstep for 3 years now and i have been really happy. Suddenly i wanted to change my micro stepping on the Z drive to get better performance. There is a bug and i have tested over and over why i have no idea but i am working on it . I have compiled the asm file in MPLAB both versions with NO TIME OUT you will NOT be able to micro step period. Only 1 diffrence between the 2 files i cant work out why just yet. Anyone wanting to micro step will have to use the files with 5min time out. I want to get the time out on my Z drive up to about 30 min and when i sort it out i will post. As you all know the time out software you stand a chance of loosing your zero. I guess Alan did not get enough donations from all including myself to sort the problem out considering how many people use his board. So lets be gratefull anyway.
Ok i got it sorted 35min is long enough for me, With permision i will post the hex and asm files due to copyright but here is the line you have to modify
incf timeout_reg, f ; Increase the timeout reg value
btfss timeout_reg, 2 ; Check we've been around the 4 times of the 16 bit
Change timeout_reg, f
ibtfss timeout_reg, f ; Check we've been around the a lot more times
timeout_reg, f (f) gave me a time out of 35 min and that is fine with me i hope this helps you guys. Have fun & happy CNC lol i mean Xmas lol
I've gotten a few donations from people (thanks! you know who you are), but I got a lot more nasty people who'd rather be demanding and unappreciative than civil.Quote:
Originally Posted by dalescnc
Makes giving stuff away for the love of it a little hard.
-A.
P.S. I'll patch the files on fromorbit.com soon. The PICStep page really needs an update.