slodat wrote: ↑Thu Jan 06, 2022 8:18 pm
DMM was of no help to me. And, when I was going through this many folks on here kept telling me to use the screw terminals. I went to great lengths to diagnose this, develop a meaningful test, and then make the change and see the improvement. I don't care what you do. I do know that for me, on the systems I've built, using the DB25 step and direction signals, as I have documented in my thread(s) about this has worked flawlessly.. going on three years.
DMM has refunded my money for their cables, twice. I won't make the mistake of buying the cables again. Anyway.. that's my experience.
Do you happen to have the color code / pin out for the cables you provided a link too?
Thanks Bob
(Note: Liking will "up vote" a post in the search results helping others find good information faster)
On page 7 is a picture of the differential driver board developed by KENTONSJ. I contacted him to ask if he would sell them and he said he would. His kit included new cables to go to the DYN4 servo drivers. I returned the DMM cables as many others have. KENTONSJ's solution has worked perfectly!!!
If you look at the first page I started with a situation where I was unable to cut concentric features - at the time I thought that maybe the parts slipped in the clamps but after several tunings, calibrations etc. I used the DMMDRV app to see I was losing steps.. I did several testing cycles on each axis and found that after each cycle I did not return to zero (steps) I was a couple thousand off and it would be cumulative meaning the more cycles I did the further off it would get. The differential board is designed to eliminate noise and that what was causing my trouble.
I'm a new kid on the forum contemplating using dyn2 drives and motors for x,y and z axis and a dyn4 drive and motor for the spindle on a taig mill upgrade. First of all, thanks for all the information shared on all the hard work you guys have done. It will be very helpful in putting together schematics and control box design for my upgrade.
I have a masters degree in EE with experience in low noise amplifier design and have a bit of prior knowledge in this area. It would appear to me that there are two sources of noise which are not being identified as separate issues. They are:
1) Radiated noise (EMI). - A differential line driver such as has been described is an optimal way of designing around that problem. A different cable which is better shielded is a less effective but still helpful solution.
2) Power supply noise. - A circuit can pick up noise through the power supply. As the devices using 24V pull more and less current, they can inject noise into the 24V power supply line. The advantage of a 24 volt circuit is that it takes a much bigger noise spike to go through the threshold of being interpreted as a false 1 or zero. Dropping the voltage from 24V to 5V with a simple resistor is problematical because you couple those noise spikes to a 5 volt system which is much more sensitive to those spikes. Electronic circuits have something called "Power supply rejection ratio" which is a measure of their ability to reject noise or fluctuations on the power supply. It is why good practice is to use a voltage regulator circuit rather than a simple resistive divider.
Possible things to try is to use the header connections, but skip the 24 to 5 volt resistor and just hook it to a solid 5V source. Use good shielded db25 cables with short runs of the wires where they leave the shielding. There really is no reason to have any skipped steps, so if testing shows them there would appear to be several solutions available.If I can put my order in for the differential driver, put me down for one if it costs less than $50.
(Note: Liking will "up vote" a post in the search results helping others find good information faster)
gregfrmatl@aol.com wrote: ↑Thu Apr 07, 2022 8:47 am
If I can put my order in for the differential driver, put me down for one if it costs less than $50.