CS-Lab Support Forum for CNC Community
Help to run this brand-new forum and stay with us.
Ask your questions, we are here to help!
Z axis loosing steps on a MAHO MH 600 T, SimCNC and delta drives
Hello eveyone,
I'm looking for some help to fix a problem on a MAHO MH 600 T for which I bought a complete retrofit kit based on SimCNC, CSMIO IP-A and delta drives and motors from you reseller in Germany nearly a year ago and since then I've been struggling to make the machine work properly again. I'm asking for help here because, unfortunately, the reseller after sale support is very poor and it takes weeks (if not months), to have a reply from them.
I must say I was very happy with the entire system once I finally got it up and running until I discovered that the Z axis loose 0,01 mm every time it retracts up and then go down 10 mm. Not a great problem with simple jobs, but it means entire millimeters in a job with lots of retactions. None of the other axes has the slight problem. At first I thought it was due to the weird gear ratio that the axis has (something like 3,27272727....), so I've change the motor gear to bring it to a decent ratio (3,6), but the problem remain and more time was lost. At the time being I've bought also the converter "sinus to ttl" to use the SimCNC with the machine glass scales, but it didn't work so I sent one of the glass-scale and the converter to Germany and there they are sice the beginning of October and I can get no answer at all.
The machine I'm taliking about is quite old in age but not in use: was bought ages ago with less than 1k hours work, now the counter show ~9k hours of which 2/3 were spent in writing programs on the old Heideinhein 332 and in letting the machine turned on 24/7 during the winter and the wet season (it had a faulty indramat drive that didn't like the bad weather).
I think there is no game in the screw as all the mechanics were meticulously checked 6 year ago and it has been rarely used from that time, then, nearly two year ago, the indramat give up completely and we decided to retrofit it.
The servo drives have all the same parameters, except the brake control on the Z axis.
This is all I think, can anyone help me on figuring out what should I do now?
Cheers
Hello,
You need to compare 3 values during two tests.
1) The simCNC position counter with the delta drive encoder counter.
2) The drive encoder counter with a dial gauge applied to the Z axis.
How to do it: Save the initial state of the values and do several dozen movements of the Z axis.
Return to the starting position and read the values.
This test will help determine where the position is being lost.
Come back to me with the test results.
Regards, Wojtek
Hello,
thank for your kind reply. I'll do the suggested tests and let you know.
Regards
Hello Wojtek,
I was looking at ASDASOFT to find the drive encoder counter, but I've got a little lost: have I to check the values in "Status Monitor - [3] The number of motor feedback pulse (Encoder unit ...)", or should I look somewhere else?
Best Regards
Look for it again.
Delta drives can display the encoder position on their display.
Hello Wojtek,
thank a lot for your support.
Following the results of the tests (hope I did it right):
| #1 | #2 | |||
| Start | End | Start | End | |
| SimCNC pos. | -138,33 | -138,33 | -138,33 | -138,33 |
| Delta Drive (Fb.PUU) | 582NN | 582NN | 582NN | 582NN |
| Dial Gauge |
| #3 | #4 | #5 | #6 | |||||
| Start | End | Start | End | Start | End | Start | End | |
| SimCNC pos. | -148,33 | -148,33 | -148,33 | -148,33 | -148,33 | -148,33 | -148,33 | -148,33 |
| Delta Drive (Fb.PUU) | 418NN | 418NN | 418NN | 418NN | 418NN | 418NN | 418NN | 418NN |
| Dial Gauge | -0,01 | -0,32 | -0,32 | -0,64 | -0,64 | -0,64 | -0,64 | -0,64 |
| #7 | #8 | ||||||||
| Start | End | Start | End | ||||||
| SimCNC pos. | -148,33 | -148,33 | -148,33 | -148,33 | |||||
| Delta Drive (Fb.PUU) | 418NN | 418NN | 418NN | 418NN | |||||
| Dial Gauge | -0,64 | -0,62 | -0,62 | -0,62 |
#1 From Z +10 to Z +20, 40 times, F500, no dial gauge; NN stands for numbers that keep changing from ~00 to ~20 (58202-58214-58219...), while the axe was still.
#2 Same as #1
#3 From Z 0 to Z +10, 40 times, F500, with dial gauge; NN from ~40 to ~60
#4 Same as #3
#5 From Z 0 to Z +1, 40 times, F500, with dial gauge; NN from ~40 to ~60
#6 Same as #5
#7 From Z 0 to Z -1, 40 times, F500, with dial gauge; NN from ~40 to ~60;
#8 Same as #7
From the datas above it seems that both SimCNC and Delta Drive keep to their values, while the machine is phisically loosing in hight (maybe it has that screw game or something mechanical...).
Following the datas from a little test: HOME to Z0, repeated 4 times
| HOME | Z0 | HOME | Z0 | HOME | Z0 | HOME | Z0 | |
| SimCnc | 0 | -148.33 | 0 | -148.33 | 0 | -148.33 | 0 | -148.33 |
| Fb.PUU | 1.7.2NN | 625NN | 1.0.6NN | 691NN | 0.2.2NN | 775NN | 8.7.0NN | 926NN |
| Dial Gauge | +0.08 | +0.06 | -0.02 | -0.03 | ||||
Z0 it's meant in the offset where the dial gauge was "nearly" at zero point (as you can see it's impossible to take a zero point and repeat it)
Meanwhile the reseller has contacted me back and they are working on the glass scale (hurra!)
Best regards
Fabio
>> From the data above it seems that both SimCNC and Delta Drive keep to their values,
>>> while the machine is physically loosing in hight (maybe it has that screw game or something mechanical...).
I have exactly the same opinion. Please check:
- Whether the Z-axis bed is not working too tight - When the Z-axis bed is working too tight, it cannot fall under the influence of its weight.
- Whether all the gears are tightly mounted - We had such a case that the gear had clearance and turned on the ball screw only in one direction (the error increased in only one direction).
In this case, just use nail polish and make a line on the connection of the ball screw or the servo motor shaft and between the gear.
If the varnish cracks at the connection, it means that the gear is not mounted tightly on the shaft.
By the way, BRAVO, you understood what the test was about and carried it out.
One more question: do you have the index signal homing option enabled?
Thank you
The system didn't accept my previous post so all I wrote was lost...
Anyway, I attach my SimCNC config screens
I'm working on the other checks: is quite difficult to reach the gear connected to the endless screw as it's positioned under the machine
Remember how I asked if you turned on homing using the index signal? I asked because I saw the non-repeatability of homing in your measurements. Turn on this option, and the precision of axis homing will increase unimaginably.
>> Meanwhile, the reseller has contacted me back and they are working on the glass scale (hurray!)
Don't be so happy!
Using a scale without removing the physical problem is just a plaster on the wound.
I mean that when you use a glass scale, CSMIO/IP-A will react with a jerk when there is a backlash.
Backlash, in the case of scales, is a sudden change of position to which the PID regulator must react suddenly.
It's like an annoying knock in the suspension of your car, as long as you drive calmly and on smooth asphalt you don't hear anything, but one bump is enough and you hear a knock, knock.
Don't be so happy!
Using a scale without removing the physical problem is just a plaster on the wound.
What a pity...
I was hoping on that as I'm not able to find any evident backslash, and, anyway, with the old system it had no problem at all of that kind (and it was based on standard DC Motors, not steppers), so I thought it would has worked.
Turn on this option, and the precision of axis homing will increase unimaginably.
Thanks a lot! I didn't know what that option means as is not mentioned in the "simcnc quick start" (EDIT: sorry, I find it), so I didn't flag it. Is there any detailed manual, other than the quick start?
Many thanks again for your time, I'll do some other tests and give feedback if I figured out something new
Regards
PARTNERS:
USA
Germany
Slovenia / Bosnia
Spain
South Africa
The Netherlands
Portugal
Greece
Hungary
Bulgaria
Kenya
Egypt
China
Serbia
Italy
Denmark
Finland
