Dyn2 losing EEPROM settings

Quote:

---

Originally Posted by mactec54

matth

Thats good you found something, you did not say anything before that you had started low with your settings, & built up to were you were at, as rowbare has said having the acceleration set so high does not achieve anything, you loose smoothness, the higher you go up with the acceleration, because Z & Y axes, are not doing very much work, if your high settings work for you, it is your machine & you can set it to what you are happy with

Having the fuse holders doing what they were, would of also, been very hard on the Servo Drives, it just show how good they really are

Rowbare his system is running a Smooth Stepper so does not really have to worry to much about Mach 3

With his Dmm System he can set everything to the max, that his Drives can handle without having a reset, he can then adjust to suit what his machine can handle

---

Sorry, probably should of said that the system tuning was done a while back before the problems manifested.
In the beginning I didn't have any problems with the fuse holders so the machine was running well (apart from forgetting settings).

The accell settings are weird because you guys are talking like my settings are through the roof but at 2000 the response is really sluggish and The Mach motor tuning screen shows that it takes almost full second to ramp the axis up to speed.

Thanks for all the help and input.

- - - Updated - - -

Quote:

---

Originally Posted by mactec54

matth

Thats good you found something, you did not say anything before that you had started low with your settings, & built up to were you were at, as rowbare has said having the acceleration set so high does not achieve anything, you loose smoothness, the higher you go up with the acceleration, because Z & Y axes, are not doing very much work, if your high settings work for you, it is your machine & you can set it to what you are happy with

Having the fuse holders doing what they were, would of also, been very hard on the Servo Drives, it just show how good they really are

Rowbare his system is running a Smooth Stepper so does not really have to worry to much about Mach 3

With his Dmm System he can set everything to the max, that his Drives can handle without having a reset, he can then adjust to suit what his machine can handle

---

Sorry, probably should of said that the system tuning was done a while back before the problems manifested.
In the beginning I didn't have any problems with the fuse holders so the machine was running well (apart from forgetting settings).

The accell settings are weird because you guys are talking like my settings are through the roof but at 2000 the response is really sluggish and The Mach motor tuning screen shows that it takes almost full second to ramp the axis up to speed.

Thanks for all the help and input.

Quote:

---

Originally Posted by matth

The accell settings are weird because you guys are talking like my settings are through the roof but at 2000 the response is really sluggish and The Mach motor tuning screen shows that it takes almost full second to ramp the axis up to speed. Thanks for all the help and input.

---

No your settings are fine, it really is what suits, your machine, every machine can be quite different, how they perform & what settings are needed, it's always hard, to know were to start, when you don't have hands on

When you get used to your machine, you may want to revisit your settings, & do some more changes, to see if you can even get it better than it is

The machine was shut down last night after a 2 hour cycle. After fixing the issue with the fuses it ran with no problems at all.

Today I turned the machine on and the X axis driver had forgotten it's settings!

Gain had gone from 65 to 16, mode had changes for Pls/dir to Rs232, onpos changed from 127 to 3 and gear number from 500 to 4096.
Trouble is I didn't know until I homed the machine and the gantry skewed faulting the y drive.

After setting the values back to normal the machine is running fine.

Don't know what to do now!

- - - Updated - - -

The machine was shut down last night after a 2 hour cycle. After fixing the issue with the fuses it ran with no problems at all.

Today I turned the machine on and the X axis driver had forgotten it's settings!

Gain had gone from 65 to 16, mode had changes for Pls/dir to Rs232, onpos changed from 127 to 3 and gear number from 500 to 4096.
Trouble is I didn't know until I homed the machine and the gantry skewed faulting the y drive.

After setting the values back to normal the machine is running fine.

Don't know what to do now!

Now my A axis driver has started to be sluggish, connecting in a checking it is reporting an over heat condition.
This is after just a 1 hour light job. None of the other drives are complaining and this drive ran for four hours solid yesterday without a problem
solid yesterday without a problem.


(Ignore me saying X driver, I meant A driver):
http://youtu.be/egj07lfodio

matth

I have asked this question before what Breakout board are you using, & how do you have it wired, It seems like you have some problems in this area

Quote:

---

Originally Posted by matth

Now my A axis driver has started to be sluggish, connecting in a checking it is reporting an over heat condition.
This is after just a 1 hour light job. None of the other drives are complaining and this drive ran for four hours solid yesterday without a problem

---

Thats an overload condition, you may have the (2) Drives Motors fighting each other, check your Gantry & Ballscrews, lower your acceleration settings in Mach

Quote:

---

Originally Posted by mactec54

Thats an overload condition, you may have the (2) Drives Motors fighting each other, check your Gantry & Ballscrews, lower your acceleration settings in Mach

---

Excellent call Mactec, when the X axis driver lost it's settings this morning the machine tried to home and then the A axis driver tripped out.
This motion has actually twisted me gantry on it's X axis rail blocks, the gantry has then been squared up by the home switches and the A axis has been leaft fighting the spring on the gantry.
To be fair to this kit its amazing that it was moving at all.

I'll square the gantry back to normal and I'm sure this will clear this issue.

I am using a very simple little breakout board connected to the ESS purely to provide some isolation / line driving capabilities.
The enable lines from the drivers are all tied together to ground via a NC relay which is only opened when Mach3 is active.
All the grounds from the drives are connected together to this board and then Step and Direction are connected to the appropriate pins.

Attachment 251794

This whole issue highlights the fact that I need to be using the Alarm outputs through some logic to initiate an EStop.

I didn't go with the DDM board because it doesn't really support a dual drive gantry arrangement. Yes it has X and <X but that is hardware mirroring of the X pulses so doesn't support homing of each gantry axis.
And if you use the A axis this disables other features of the board.
I was using a USB BOB which worked very well until I found out it didn't support axis slaving hence the change to the ESS and this simple BOB.
Of course now I have the ESS I could use the DMM board and other BOB to provide additional IO.

I'd be happy to change out this board If someone could give me a reason why it might be causing issues. When I reported the memory loss problem to DMM their response was "it's a known software issue, we'll send you new drivers".

Powered on this morning and the X axis driver had again forgotten settings (Torque a Filter cons and On position range). I am now as a matter of course plugging in with DMMDrv to check each drive before I start Mach3.

I am going to be powering the machine on and off a lot today and monitoring the drives. So far I've only seen the new X axis driver doing this.

No response from DMM so far on this issue with the replacement drive.

- - - Updated - - -

Powered on this morning and the X axis driver had again forgotten settings (Torque a Filter cons and On position range). I am now as a matter of course plugging in with DMMDrv to check each drive before I start Mach3.

I am going to be powering the machine on and off a lot today and monitoring the drives. So far I've only seen the new X axis driver doing this.

No response from DMM so far on this issue with the replacement drive.

matth
There is something going on with how you have wired your system, that is causing this problem,if the Drives were going to do this they would of done it the first time you turned them on

It is hard to troubleshoot, when your system has not been wired to Dmm specification, others have not used there Breakout Board before without any problems, I think you have something going on in this area that is causing your problem with the Drives

My be how you have wired your Drive Enable also not using there Drive Alarm, some of the Drive Safety features are not being used, which are there to Protect the Drive

With the Alarm being by Passed you do not have any warranty with there System

I can't agree with that assessment.

All the drives are wired the same yet I was only seeing the mem loss problem on two of the old drives and I'm only seeing it on one of the replacement drives.
The fault is on startup, before Mach3 or the ESS have started. If the software fault was going to manifest itself the moment the drives were turned on how did it ever get past QC???
These drives are wired as per the specifications laid out in 3.6.3 of the Specification_DYN2.pdf. Yes, I am not using the DMM BOB but there's nothing in the DMM documentation that mandates this.

The Alarm output from the drive is just a simple output pin that goes high when the drive goes into a fault condition. The drives that have lost settings are not going into a fault mode so I can't see what one thing has to do with the other. I also don't see how not using the Alarm out would somehow cause the drive problems or invalidate my Warranty.

If the drive wiring was causing the fault then surely all the drives would exhibit the fault and it would be a regular and repeatable issue.

If someone could point out how my system is not wired to specification and also how external wiring can affect an internal EEPROM I'd be more than happy to remediate it in any necessary way.

Can I also remind you that DMM directly told me that the loss of settings was a known software issue.
This was after I sent a mail describing the issue, I was not asked about wiring or told to check anything.

I would very much like to hear back from DMM on the issue of the replacement drive also losing it's settings. If they can offer any suggestions as to how my wiring is bad (our you for that matter), I will gladly listen.

FYI, I know how an EEPROM works. How is my external wiring causing an issue that is getting all the way past all the other drive electronics and then causing the EEPROM to erase or modify certain values?

matth

I must remind you that the you had the 48VDC wiring incorrectly done,Which you have changed/corrected, I believe you have the same problem as well, with your Breakout Board to the Drive wiring also

A spike in the switching on, the way you have it wired will Zap the Drives, how it affects the Drives is always a little different, Dmm try to protect against this kind of thing, but when people do there own thing, with wiring, anything can happen, Your Enable should not be wired as you have it, this is not to Dmm's specification

If you wanted to Switch anything it should of been switched at the input to the Power Supply that supplies the power to the Drives

If the Dmm Alarm system was in place, you would not of been able to do a Reset in Mach, then your machine would not of run/crashed as it did, because the Drive was in a fault condition

Quote:

---

Originally Posted by matth

Powered on this morning and the X axis driver had again forgotten settings (Torque a Filter cons

---

These ( 2 ) settings /Parameters are not active & it does not matter if they are there or not, it was used for RS232 & analog connection only, Plays no part of your Step/Dir control

OnPosition is important though, if you had an EEprom Problem, it would affect all settings not just ( 1 ) setting, that is why you have a problem in what you are doing in your system

How does my earth point get a "zap" at power on? It's not the neutral bar. That would mean I had massive leakage current via the CPC and the RCD I have on this circuit would trip, which it never has.
I have removed all low voltage earth returns from my main earth bond point and installed a earth busbar (separate from all 5v earths).
However this is not common practice, please lookup earth bonding where the recommendations are to bring all system earths back to a central point, which is how my system was wired. Even with a DMM bob in place this would be no different, how do DMM protect against this?

Please show me the specifications from DMM as to how my enable lines should be wired? What is the problem with tying them to ground until activation?

What am I switching after the PSU's, the wiring goes from the PSU, into a busbar them through a Breaker then into a filtering cap then into the drives. At no point is it switched!

My machine didn't crash and the drive didn't have a fault, I started the machine and the X axis driver had forgotten it's settings, without knowing this (do you connect to every drive at every power on and double check it's internal settings?) I asked the machine to home.
This twisted the gantry because the X axis wasn't driving properly whereas the A was. The A then faulted. After this I reset the system but didn't realise the gantry was now physically twisted.
At no point would having the alarm outputs connected of changed what happened.

The fact is the original fault was explained to me by DMM as a software issue and one of the new drives is acting just the same.
Yes I have had some other issues with power wiring which have been sorted and as far as I can see are unrelated.

How do you know that a software problem that you admitted to not knowing about would affect all settings?

Quote:

---

WARNING!:
Never use serial communication to set the Drive cons or configuration all the time in a rate of certain time period, that
will cause EEPROM busy in writing all the time and short life, the guaranteed writing cycle for EEPROM is 1 million times.

Once a group of cons and configuration are set, try to use it until next necessary parameter change.
3.1 Read servo parameters out from Drive before revising them
The Drive will use the default servo parameters or constants for the first time power on reset(POR), that default parameters will
ensure the servo is at best performance for no inertia load and stable for some how small inertia load.

---

Is that supposed to mean, change one parameter at a time, saving between each setting?
Or does it mean, don't continuously save the settings at a high rate of speed (how would this be possible?)

Also no mention of what "first time power on reset" means. It's safe to assume a new drive would have the defaults stored,
but maybe the "first time power on reset" is occurring somehow and defaulting the eeprom?

If you have a graphing voltmeter, can you measure the voltage at the drive power input for fluctuations?
It seems possible that a rapid power on / power off sequence could cause the drive to lose it's settings.
Perhaps your DC supply was damaged somehow by the fuse's intermittent connection? (I know, not likely)

Good points Alan but I don't see that it's the defaults as it seems that random settings change to random values. I've seen main gain change from 65 to 24 and 16, onsposition from 127 to 0 and pod 3 etc etc.

I suppose one of the PSU could possible behave like you describe bit I have 3 in parallel so surely it one was fluctuating then the others would fill in. I can't see them going overvolt as they are fully regulated.

I have a scope but it's not a storage scope so I can't capture any data.

Quote:

---

Originally Posted by alan_3301

Is that supposed to mean, change one parameter at a time, saving between each setting?
Or does it mean, don't continuously save the settings at a high rate of speed (how would this be possible?)

Also no mention of what "first time power on reset" means. It's safe to assume a new drive would have the defaults stored,
but maybe the "first time power on reset" is occurring somehow and defaulting the eeprom?

If you have a graphing voltmeter, can you measure the voltage at the drive power input for fluctuations?
It seems possible that a rapid power on / power off sequence could cause the drive to lose it's settings.
Perhaps your DC supply was damaged somehow by the fuse's intermittent connection? (I know, not likely)

---

As per your first question, the documentation seems a bit limited in this area. As you say, how can you save the params so fast as to lockup the EEPROM?

so is it down to 1 drive screwing up the parameters now?
If so, it sounds like it's time to swap with another working drive in the panel.

if the problem follows the drive, it probably has to be sent back.

With my luck, the problem would go away for a couple months until I had forgotten about the problem.