ELS Conversion on Weiss 1127

I made some progress mounting the scale for the X axis on my lathe. I spent a bunch of time trying to figure out the best place to mount the scales, and after some debating with myself i decided to place it above the cross slide gib adjustment screws and with the reader towards the back of the saddle. With the cover on i shorten by tailstock travel by less than .5" so that's not terrible. Drilled, tapped, leveled and the scale mounted to the bracket. Here's what it looks like:

PXL_20230107_204347192.jpg

I wasn't sure if it was a good idea to mount the read head with only one bolt and while it might have been OK i decided to make another part. Well truth be told i made a new part twice! In the layout i flipped the part around enough times that i ended up laying it out so that the leg was on the wrong side, Darn it!

IMG_20230108_132714_02.jpg

Re-made the part with the leg on the correct side.

PXL_20230109_000607376.jpg

Got as far as center punching the holes for the read head screws but no drilling or tapping.

PXL_20230109_000614255.jpg


The pieces are aluminum and look way rougher than they feel. I'm an amateur at best so i'm pretty thrilled that i've hit my tolerances so far.

Don't know if i really had to remake this piece as it seemed affixed quite firmly but i didn't want to have to pull the lathe away from the wall, take off the splash guard and potentially replace the reader if it gets broken in the future. I did find out that after all of the comparisons on AliExpress i did end up with 5 Micron scales for both the X and Z. I had thought that the X axis was 1 Micron as it was a slim line model built for the cross slides but i attempted to purchase both scales from the same seller and clearly didn't pay enough attention when ordering.

Making progress, slowly but surely.
 
Major progress with the scales today. The last time i posted it was the third time i made the part and today when drilling the holes for the encoder reader mount i broke a #29 drill bit. I wasn't pushing it hard or at least i didn't think i was but it snapped off flush as it was about all the way through the part. Pretty sure it loaded up with chips and then galled up in the hole. I tried a few things to get the broken off bit out of there but in the end i ended up drilling it out with a annular cutter and then turning a plug down and pressed it into the hole. You can see it in the pic but i didn't really want to make this part a fourth time. Here's a pic of the scales mounted to the cross slide:

PXL_20230122_002436334.jpg

The Z scale was much easier to mount and went much faster.

PXL_20230122_002559901.jpg

I have yet to install the covers on either scale. If you look closely you can spot some mistakes. I want to point out that the holes don't line up on the angle bracket from the factory that connects the Z axis encoder reader to the carriage. The slots are large enough and spaced such that you have to 2 washers for each screw and use one inner and one outer screw hole. Not a big deal but it must be a universal bracket that "fits" in the bare minimum way possible.
 
Last edited:
May I please ask which ELS system you are adding?
 
@WobblyHand - RE:troubleshooting the scales - i am getting the same type of results on both sets of scales. On the X axis i connected a dial indicator and measured .100" which shows up as .200" on my dials as they are direct read. The ELS DRO read .190" if i move it really slow which sounds decent but once i wind it in and back out at a faster rate it doesn't match the indicator. I don't think it is this scale as the Z axis does the same thing. The rate i'm using is no faster than the DRO on my mill. It almost seems like i'm missing counts on the encoder or something - sometimes it also doesn't increment. Before mounting i plugged the Z axis into my mill DRO and it seemed like it read fine although i didn't do any scientific measurements to verify.
 
More testing shows that if i turn the X axis very slow it will skip between .001 and .000 and until i spin it fast or reset it the display will not update.
 
@WobblyHand - RE:troubleshooting the scales - i am getting the same type of results on both sets of scales. On the X axis i connected a dial indicator and measured .100" which shows up as .200" on my dials as they are direct read. The ELS DRO read .190" if i move it really slow which sounds decent but once i wind it in and back out at a faster rate it doesn't match the indicator. I don't think it is this scale as the Z axis does the same thing. The rate i'm using is no faster than the DRO on my mill. It almost seems like i'm missing counts on the encoder or something - sometimes it also doesn't increment. Before mounting i plugged the Z axis into my mill DRO and it seemed like it read fine although i didn't do any scientific measurements to verify.
Can you show a picture? Not visualizing the issue(s) yet. If the DI shows 0.100", what does the ELS display show for that axis? 0.1900?
Is the display stable? Not flickering? I haven't tried this on any other scales. I suppose I could try it on my mill DRO scale.

Not saying this is your issue, but I have had glass scale issues that appeared to be missing counts, or in my case a difference between going one way vs the other. I called it the non-return to zero. Move 3". Move back 3", the answer should be zero. In my case, I cleaned the inside glass and it made it a little better. But it didn't fix it. A replacement read head fixed it.

It's possible that there's an incomplete connection or even a wrong one still. Is it possible to take a picture of the incoming DRO cable with the wires colors exposed with the cable hood off? So we can see the solder joints and possibly the connector pin numbers? And maybe a picture of the inside of the ELS box, so we could do a visual comparison of pin positions and or colors?
 
More testing shows that if i turn the X axis very slow it will skip between .001 and .000 and until i spin it fast or reset it the display will not update.
This sounds like a missing connection, or something like that. Can you take some pics of the DRO cable end GX12-4 in enough detail to read the numbers and to see the solder joints? And maybe open up the ELS box and take a picture of the DRO cables?

Edit: It's possible my system is internally consistent, but off by a pin. :( This is the reason for opening the box and doing a visual on the wires both inside and outside of the box.
 
Last edited:
Welp, not a bad connection but worse; a bad scale. I confirmed this by re-soldering the DB9 connector back on and then plugging into my mill DRO. The read head is not right - i don't know if i damaged it during installation or if it came that way. That's the slim scale and if i have to purchase another one i might get a 1um scale instead.

The Z scale seems to read reliably but is off by a factor but at least it's a consistent factor. Using a 1 2 3 block it reads 5, 10 and 15 inches. I moved the DRO until it read 25" and then double checked quickly and it was 5" of actual travel so it's off by 5:1.
 
Welp, not a bad connection but worse; a bad scale. I confirmed this by re-soldering the DB9 connector back on and then plugging into my mill DRO. The read head is not right - i don't know if i damaged it during installation or if it came that way. That's the slim scale and if i have to purchase another one i might get a 1um scale instead.

The Z scale seems to read reliably but is off by a factor but at least it's a consistent factor. Using a 1 2 3 block it reads 5, 10 and 15 inches. I moved the DRO until it read 25" and then double checked quickly and it was 5" of actual travel so it's off by 5:1.
That's pretty good news on the Z scale! That's just a calibration factor. My scales are 0.001"/div. Yours are 5um, which is 0.00019685"/div.
There's about a factor of about 5, which you measured. Basically, I had 1/1000 as the cal factor. You will need 1/5080, or very close to it. I will put in a #define in the code for different scales to make this easier for users to select.

Bummer on the X scale. I'd bet it is the read head. However, I'd try to get a full replacement. If you can't get a whole new scale, then settle for a read head. Or order a 1um scale. It's better to have 1um on the cross slide.

Well, some ups and some downs. But making progress! Hope you get the X scale sorted soon.

Hook up your stepper, that's really a fun moment when you start to see stuff move. Since the tach is working on the ELS (except for direction) the encoder is running. The tach uses the encoder information, so we know the encoder works.
 
Back
Top