[4]

DRO drifting

[3]
[10] Like what you see?
Click here to donate to this forum and upgrade your account!

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
Hi all,

I was in the workshop yesterday happily turning away when I noticed that my DRO had drifted. Going back over the parts I was making it suddenly shifted by 1.3mm diameter after about 2 hours of machining. The drift was sudden, i.e. all dimensions were fine then the last three diameters were all out by the same amount (approx. 1.3mm). Thankfully I noticed as I must admit I have become quite reliant on the DRO as up until today it has never been out by a significant amount.

I do have a theory that it was the magnetic base which I put on the cross slide next to the scales. It's a glass optical system so I would have thought that the mag base shouldn't cause an issue? The timing does coincide with the drift however I couldn't replicate the drift by placing the mag base in various locations on and around the scales afterwards which has left me rather puzzled.

One thing I have also noticed is that it can drift when the machine is not in use. Because of this I always check before each use but I'm still puzzled as to why it has been doing this. In this instance there wouldn't be a mag base anywhere near the scales.

So I'm wondering if anyone else can suggest a reason as to why my DRO drifts?

Thanks,

Sam
 

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
Thank you for the suggestion. That's certainly a possibility and although I didn't notice anything at the time it is very possible even if I had seen the lights dim I wouldn't have thought anything of it. Thinking about it, it would also be worth me looking where the leads hook up to the display unit as I was raking behind the lathe yesterday too.
 

TonyRV2

Active Member
Registered
Joined
May 6, 2018
Messages
58
I'm assuming that its drifting on just one axis. Swap the cables around on the back of the readout. If the problem stays on the same axis, you know its the readout itself and therefore an electrical/electronic problem. If the problem changes axis then you know that its an issue with the scale on the machine. This will at least get you looking in the right place for the problem.
 

homebrewed

Active Member
Registered
Joined
Jul 28, 2017
Messages
212
Check the simplest stuff first -- the DRO mount points. I had a similar problem and was blaming the scale for a troublesome glitch in the displayed position. It turned out to be a loose mount for the scale. It has very small machine screws and I was overly concerned about stripping the threads, so they weren't nearly tight enough. They eventually worked loose and the problem began.
 

pontiac428

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Apr 23, 2018
Messages
395
Perhaps a power dip?
Would probably be worth your time to measure voltage drop during motor startup at the point where your DRO connects to the mains power.
 

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
Thank you all for your suggestions. Certainly a few things for me to junk about!

I’ve only noticed it on the X axis however I zero the Y axis so often that I probably wouldn’t notice if it drifted overnight.

The voltage drop on lathe startup is a good suggestion. Certainly it is enough to din the lights although the DRO beeps when powered off and t never does that upon starting the lathe. Still, I suppose it could cause it to get confused.

I have checked all of the mounting points for movement and there is nothing as far as I can tell. Certainly if I try to love the scale by hand I get nothing until the cross slide moves to take up backlash.

One thing I did notice today was that the earth cable wasn’t connected at the back of the DRO unit. I remeber now that when I installed it the earth cable supplied was only a few inches long and therefore utterly useless. I had intended to fit a longer one but clearly it slipped my mind. I’m not sure if this could cause any issues such as this?

I’ll have a go at measuring voltage drop over the next few days. I would do it this evening but I have the furnace running at the moment and the supply is only rated at 16 amps to that side of the workshop which means it trips easily when the lathe fires up. I should probably run a spur from a different circuit to power the furnace!

Thanks again for all of the suggestions! It’s given me a few things to investigate and without your helpnid probably still be scratching my head!
 

kd4gij

Active User
Registered
Joined
Feb 7, 2011
Messages
4,769
Are you talking drifting over night or drifting while in use? If over night, then temperature change will do that.
 

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
Both. It has always drifted overnight but that has never been of a huge concern until Sunday when it drifted whilst in use. I had wondered whether the two issues could be related. I couldn't tell you exactly how much it can drift overnight but it is often fairly significant (i.e. more than a few mm) so I'm not sure different rates of expansion between the cross slide and the scales would be the cause of that.

That got me thinking though – how much will temperature changes affect the DRO?

As an exercise I calculated the change in length over a 10°C change in temperature using the formula L1 = L0 x α x ΔT

Where L1 is the change in length, L0 is the original length, α is the coefficient of thermal expansion for a given material and ΔT is the change in temperature.

A quick google and I found the coefficient of thermal expansion of cast iron (the cross slide) to be 10.8 x 10-6 m/m K (rounded to 11 x 10-6 m/m K) and aluminium (the scale housing) to be 21-24 x 10-6 m/m K (where I took 24 x 10-6 m/m K as a worst case scenario).

So to plug the numbers in we have for the scale

L1 = 0.2 x 0.000024 x 10 = 0.000048m = 0.048mm

And for the cross slide

L1 =0.2 x 0.000011 x 10 = 0.000022m = 0.022mm

Which gives us a worst case difference of 0.026mm (0.052mm across the diameter).

This wouldn’t be enough to cause me serious issue on the parts that I am making although I think that it is something that I will check periodically throughout the day going forwards to account for this.
 
Last edited:

TonyRV2

Active Member
Registered
Joined
May 6, 2018
Messages
58
Get that earth ground connected!! As a former automation engineer and having taught electronics for 32 years, I can tell you first hand of a number of problems associated with poor grounding, especially in an electrically 'noisy' environment like a shop. I'm not saying that this is the source of your problem, however, especially since you'd said the drift will occur overnight when most machines would be turned off. And it would be unusual for a grounding problem to affect just one axis as well. What I'm saying is that this is the source of a potential problem and is easily eliminated. Have you swapped the cables around on the back of the DRO yet?
 

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
No I have not swapped the cables round yet. I installed an earth cable last night so I plan to see if the DRO drifts overnight before going any further with the investigation. I'm not planning on using the machine this evening but I will do a test cut just to see if it has drifted at all. I'm hoping that all is well and it was the ground that was causing me the issue however if it has drifted I will switch the cables round, set a size and leave it again to see if it drifts.

Frustratingly I don't know over what time period it drifts - to be honest I've never really paid it much attention other than re-setting the size before each use. I don't use the machine every day so it could happen overnight or it could happen over the space of a few days.
 

markba633csi

Gold
Registered
Joined
Apr 30, 2015
Messages
3,463
How long have you owned this DRO system? The drift could be a sign of impending failure of an internal part, such as the internal crystal oscillator.
Also, have you tried tapping on the unit to try and duplicate the failure?
Mark
 

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
I bought this system new just under a year ago. I'm not entirely sure who manufactures it - it's of Chinese origin and is branded as a Warco unit (UK's version of Grizzly and the like I think) and it is fitted to a Warco GH1330 lathe (13" x 30"). Links below if you're interested.

I've not tried tapping the unit but I have given everything a good wiggle to see if anything is loose. So far I've not come across anything.

Since I added the ground the X axis hasn't shifted and the Y has shifted 0.015mm which could easily be due to expansion as suggested above (it has been hot in the evenings and it was cool in the workshop this morning when I checked). I'll check back again this evening and see if it has drifted further or, if it is due to expansion, it's drifted back to 0.

https://www.warco.co.uk/metal-lathes-metalworking-lathe-machine/26-gh1322-gh1330-lathe.html#/906-gh1300-gh1330_1ph
https://www.warco.co.uk/digital-readouts-scales-dro/292-digital-readout-scales-glass-optical.html
https://www.warco.co.uk/digital-readouts-scales-dro/303110-warco-digital-readout-dro-display-counters.html
 

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
Hi Guys,

A quick update for you. Apologies that it has taken a while to get back - I've been away with work so haven't had much of a chance to look into it.

After connecting the earth cable I zeroed the X and Y axis and left them for a few nights to see if they drifted and nothing (apart from a very small amount on the Y axis as mentioned in an earlier post which I am putting down to thermal expansion in the machine). Annoyingly when I got back over the weekend I accidentally knocked the cross slide handwheel before getting a chance to check if had drifted while I was away. I re-zeroed the X axis and so far nothing.

So, I believe that, with your help, I have traced the fault to the lack of an earthing cable which has now been rectified.

I will continue to monitor and post back here should anything change but I wanted to give some feedback on the issue and say thanks once again for all of the suggestions above.

Thanks again,
Sam
 

RobertHaas

Active Member
Registered
Joined
Mar 16, 2016
Messages
91
faulty ground can show up during operation on a lathe. Pretty common really as any time you rotate conductive material it generates an electrical field.

Most people never ground there machines but it is a good idea.
 

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
Finally got my lathe back after it was away for repair with the manufacturer for an unrelated issue. I’ve had a few projects lined up so I started making chips right away but a lot of my non critical dimensions were way off. I don’t usually bother checking these and rely solely on the DRO and I couldn’t understand it. It would appear the drift is back!

When trying to troubleshoot the problem I popped in for a cup of tea and came back to find that it had drifted by four and a half meters! At least that was easy to spot! Most of the time it’s by a mm or two so much harder to spot.

I switched the x and y axis and was still getting the error on the cross slide so I assumed it was the scale itself however this morning I noticed the error on the Y axis (the X axis scale has been removed now so I’m not tempted to rely on it!).

If I touch the cutting tool on a piece of barstock and zero the axis and move the carriage to the far end of the scale then back again it is rarely 0. It can be out by a couple of mm or more each time.

I have verified that the earth is actually earthed with a volt meter and all is good there but now I’m lost! Electrical issues are not my strong point!

Does anyone have any further suggestions?

Thanks in advance!

Sam
 

Cobra

Active User
H-M Supporter - Gold Member ($25)
Joined
Apr 24, 2013
Messages
680
Had a similar issue with magnetic scales. Called the supplier and described the problem.
He immediately said the reader was bad and sent a replacement.
Solved the problem.
 

RJSakowski

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Feb 1, 2015
Messages
3,862
I have had the Grizzly version of your DRO on my mill/drill for fourteen years with issues. Intermittent failures can be difficult to track down. Just when you think you have the problem solved, it pops up again.

I'm assuming that you have a two axis DRO and that the y axis is the cross feed (conventionally called the x axis) and the x axis is the carriage movement (conventionally called the z axis). You stated that you were experiencing problems on both scale but different manifestations. Your x axis was jumping to a new value and your y axis experienced a slight drift. You have swapped the cables already and stated
Some thoughts:

Two or more separate failures usually have a common cause. In my experience, unrelated coincidences are rare.

As far as I am aware, the scales do not output absolute position. They sense magnitude and direction of movement and send pulses to the readout with the counter's position registers located within the readout. The readout does retain the last readings if power is lost and reloads them when power is restored. There is a single point on each scale that is an absolute reference point. It can be utilized using the REF function on your readout.

As to thermal expansion, you really should be looking at expansion coefficients of the glass scale and the aluminum backing that it is bonded to in your analysis. Depending on the type of glass used, the expansion coefficient is between 4 and 6 x 10-6 m/m K. Changes in temperature would cause the scale to bow slightly but, since the scale is fixed at either end, not likely to create a positional change significant enough to cause a meaningful shift in the reading. Similarly for the mounting to the lathe.

In troubleshooting, try to eliminate as many possible causes as you can. I would reconnect the second scale and document the anomalies you are seeing.

Based on what you have said, I would suspect an intermittent electrical problem, a bad connection or bad solder joint. In either case, it would most likely require a return for replace or repair. Assuming that you have two different manifestations on the scales, Record the exact problem with each when it occurs, then swap connections. The problem will either go with the scale or remain which will tell you whether the problem is with the scale or the readout.

Armed with that information, contact the vendor and present your evidence. You should still be under warranty. More than likely, the vendor would replace the entire DRO package.
 

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
Thank you both for your input. It does sound like it's best to get the supplier involved so I've sent them an email with all of the supporting information. Now I just need to remember how I got by before I had a DRO system!
 

GT Mills

Mohawk Aero
Registered
Joined
Oct 31, 2016
Messages
35
I am experiencing DRO drift as well. Mine is a little different, but I will try the grounding improvements suggested here. I might try a frame ground from the meter to the mill.

My problem shows up intermittently. It is a 3-axis glass scale Chinese type, costs $300 just purchased on Amazon. First attempt to get it installed and working failed. The DRO head was faulty and completely screwed up. The vendor replaced the head. Now I am frustrated with the zero base changing. The vendor is asking if I want the scales replaced next, since I thought they may have been damaged when delivered with the first head that was ruined in shipping when I discovered this problem.

I set X, Y, Z zero. I mill some cuts for 30 minutes or so. Everything will remain OK for a while. At some point when I shut down the mill to change tools or something, the X axis changes .010 - .015" at the moment I shut off the VFD, and the readings change with the spindle speed. Then, later on, the Y axis will change also when I shut down by a few thou. As I continue to work for the next hour or so I see the Z axis has also drifted a bit. The Z scale is not attached to the mill, it is sitting on the bench because I haven't bothered to install it yet. But I keep it plugged in because the head doesn't seem to work right w/o all three plugged in.

Depending on where the mill table is positioned, I get different drifting results. At times, all I do is reduce the mill speed with the VFD and this changes the zero on the DRO head, sometimes just X, then sometimes both X & Y. This is what led me to think the scale was messed up.

It seems that the longer the scale, the more the drift.

Any comments, thoughts, ideas greatly appreciated.
 
Last edited:

SamI

Active Member
Registered
Joined
May 31, 2018
Messages
130
If this drift occurs when you alter your VFD could it be electrical noise? I must admit i know nothing of these things so can't offer any more advice but it could be worth considering.
 

GT Mills

Mohawk Aero
Registered
Joined
Oct 31, 2016
Messages
35
The vendor first sent me a new head, then sent me a third head and three scales. I now have three heads (one head is definitely screwed up internally and in a state of constant alarm) and two complete sets of three-axis scales. STILL had problems, no matter what combination I set it up with both on the mill and with the scales on the bench, the head attached or on a plastic bucket, plugged into various circuits coming off the main breaker panel, maming sure all earth grounds and frame grounds ohm-ed out perfectly! WTF? Today I did something I had not tried before: ELIMINATE the third prong (110v) safety frame ground by running it through a 3 prong-to-2 prong plug adapter. Hot dang, it WORKED!!! Who would have guessed that REMOVING the safety frame ground would get rid of this exasperating problem? SO now I have two brand new, working, 3-axis DRO's with scales, and a third DRO head that seems to have a short circuit which I believe can be repaired, all for just $300. God works in mysterious ways. Now I just need a lathe or another mill to install this other DRO kit on, haha.
 

mksj

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 12, 2014
Messages
2,095
VFD electrical noise is a common problem/issue both EMI and RFI, the degree of interference can vary and also DRO's susceptibility. Grounding and having a good electrical filter or using a screened transformer for the sockets can help reduce these problems. I typically use inline noise filters on some of my VFD installs and pay particular attention to grounding. Grounding and shielding of cables is not always straight forward, but the VFD electrical noise can play havoc with low level signals. I had an issue with a MachTach where the display went haywire if the sensor cable shield was grounded to earth, and completely disappeared when the shield was floating. A noise filter or isolation transformer may help to reduce the interference, but erratic readings in the absence of electrical noise suggests a component issue in the head unit, in particular if both scales share similar problems. I did hear of a recent issue with the earlier Easson 12B/C DRO's were there was a firmware programming error that caused reading errors form the scales.
 

homebrewed

Active Member
Registered
Joined
Jul 28, 2017
Messages
212
I've had problems with noise affecting my DROs as well. They are the capacitive type, which are known to be more susceptible to noise. I tried a line filter to no avail, and many variations on grounding, also with little improvement over the long term. The best solution for my DROs was to open the sensor boxes up and solder more capacitors between Vcc and ground (probably not a viable solution for everyone). The remote-readout design places the batteries in the display unit so the power has to travel through the cables -- which appear to act as magnificent antennae for picking up noise, even IF they are shielded.

I'll have to try the 2-wire scheme to see if that helps further reduce my noise problem. Thanks for the update!
 

GT Mills

Mohawk Aero
Registered
Joined
Oct 31, 2016
Messages
35
VFD electrical noise is a common problem/issue both EMI and RFI, the degree of interference can vary and also DRO's susceptibility. Grounding and having a good electrical filter or using a screened transformer for the sockets can help reduce these problems. I typically use inline noise filters on some of my VFD installs and pay particular attention to grounding. Grounding and shielding of cables is not always straight forward, but the VFD electrical noise can play havoc with low level signals. I had an issue with a MachTach where the display went haywire if the sensor cable shield was grounded to earth, and completely disappeared when the shield was floating. A noise filter or isolation transformer may help to reduce the interference, but erratic readings in the absence of electrical noise suggests a component issue in the head unit, in particular if both scales share similar problems. I did hear of a recent issue with the earlier Easson 12B/C DRO's were there was a firmware programming error that caused reading errors form the scales.
Thank you very much for your reply! It is welcome news to read that someone else was able to isolate & solve a similar problem by floating the earth ground instead of connecting it. It never occured to me that the problem may be tied to the VFR rather than the motor itself. I had tried a line filter, but the one I had on hand (integrated into a fancy power surge strip) didn't fix the problem, and I after that I wasn't really sure another type might help.
 
[6]
[5] [7]
Top