Height gage repair?

Sounds to me like a bad solder joint. I fixed a relatively expensive Chinese vertical gauge by replacing the (odd frequency round can) quartz crystal. Just as well, there wasn't much else in there (typical modern chip/blob construction but very nicely made metal bits and carbide tips).

I also found a bad power switch in a pair of Chinese calipers, but unfortunately it was a faulty wirebond under the blob rather than a problem with the usual conductive silicone/interdigitated PCB pattern switch (my assumption that led to cracking it open is that it might have gotten dirty or wet, but no....).

Thanks, it is possible the solder is cracked because the board is below the operating buttons and things have been getting stiff over the years. I'll have to take a look at it myself once my current tech is done with it. Who knows, he might be able to find the part or someone that has the schematic.

Ed Huesers

If the encoder is seriously scratched, consider it a write-off. I've got a couple of their 6" digital calipers (which require a no-longer-made mercury cell -- PX-13 IIRC), and there are other things which can make it not work.

Flooding it with coolant will keep the sensors from working, but spritzing with WD-40 helps greatly. Freon TF (if you can still find a can) works much better.

The other thing which can keep it from working (or cause it to error frequently) is if the two sensors are not quite aligned right. You can't see it without a microscope, but the scale (protected down in a groove in the handle) is a long quartz glass coated with metal in a pattern like this (assume that the bars and the space between the bars is equal) :

| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |

sliding just above it is another piece of quartz glass, with bar patterns just at the two ends. I'll draw it above this paragraph so you can see how it works with the main scale:

Note that one set (as shown) is half-way between the bars of the main scale, and the other set is in line with the main scale.

If you look at this with the assembly carrying the LEDs and photodiodes off, you will see that one of the two looks dark, and the other bright. Move it a half bar, and they will switch places.

The electronics watches these with the LED and photodiode at each end, and knows which direction it is moving by the which direction the one changes while the other is in which state.

If that short double-ended scale is rotated just a tiny bit it will stop working properly, so the first thing to do is to adjust it so the ends alternately go dark as it moves. Once this is done, the only other thing to do is to get the LEDs and photodiodes back in place, and make sure that the whole thing has fresh batteries or cells -- or is getting the right voltage from a wall wart, depending.

Beyond that -- the failures could be:

1) A LED dying (they are usually IR LEDs, so it is hard to tell just by looking -- except for looking through the lens and viewfinder of a cheap and old digital camera. The later ones were made with an IR-blocking filter, which makes them useless for this purpose.

2) A photodiode failing. Checking this requires being able to watch it with an oscilloscope while moving.

*But* -- if either of these failed, it would be bad enough so it would be beyond giving error displays.

3) The logic chip -- usually buried under a lump of epoxy on a tiny circuit board -- not replaceable, except if you have a spare whole board. You would not get error messages from this, either -- usually nothing would show at all.

4) The LCD display -- connected to the board by zebra strips (alternate chunks of insulating and conductive rubber, bringing connections to the LCD -- and possible to need cleaning and replacing. If the LCD itself has an open trace, you won't get segments of the numbers to light.

If the head has gibs, they need to be tight enough so the head won't tilt enough to cause problems with the bar code encoder I described at the beginning of this. First try to fix things is to adjust the gibs to keep the short bar code strip form changing angle. The pressure on the component might just be changing the angle of the encoder strip enough to cause errors.

I don't think that a schematic will be much help. The only repairable things are the angle of the encoder strip, maybe replacing LEDs and photodiodes (but your symptoms say that this is not the problem), or the entire LCD or logic card. (Remember, my description is form an early digital caliper which used the glass encoder, and I have no direct experience with the height gauge, but if they are from the same time, I would expect mostly the same construction.

I hope that you are expecting to read the newsgroup for the reply, and not hoping people will connect to the web site to tell you how to fix it.

Good Luck, DoN.

Got it Don, thanks for the info. So, my tech was able to get it working. He said the signal is getting weak and doesn't know how long it'll keep working. Hope it lasts a bit longer,

Ed Huesers

[ ... ]

"Getting weak" is a symptom of a slightly mis-aligned encoder strip.

The alignment is set by two screws holding a spring bronze strip holding the encoder strip just very slightly above the long encoder in the body. (In my calipers, it is in the groove where a rack gear would be for a dial caliper -- and when fully closed, the main strip is hidden by the depth gauge extension which rides in the same slot.)

Anyway -- loosen the two screws, then tighten them just enough to hold it in position and watch as you slide the head along.

Aligned properly, you get fully dark or fairly clear reflection depending on position.

*Slightly* out of alignment keeps it from going fully dark or fully reflective -- but enough to work most of the time, likely with a greater sensitivity to a slight tilt of the head from loose gibs. *Way* out of alignment you get Morié stripes when looking at the encoder. And it won't work at all. (All of this checked without the board with LEDs and photodiodes in place.)

Tweak the position of the encoder strip (short glass scale) by moving one way or the other and observe what changes happen. Once you have it right, alternate tightening the two screws a little at a time, so you don't disturb the position before you get the screws tight enough.

I presume that this is powered by a wall wart, not the unobtanium PX-13 mercury cells that my calipers of the same flavor used. If the PX-13s, it would be long dead for a lack of replacements of the proper form factor and voltage.

Good Luck, DoN.

Well, Thanks again, I'll save this in case I do have further problems. I got tired of paying 70/80 bucks for a battery and had an electronics tech friend of mine make up a case that accepts 4 double A batteries. I've been using that for nearly 10 years now. I've had the height gage since 1994 and love it.

Ed Huesers

Hmm ... the original PX-13 cells (which I suspect were your $70/$80 priced ones near the end of their availability) were 1.35 V (as were all mercury cells). Standard zinc-carbon cells are 1.54 V at start of life, which means that your gauge is running from 6.16V instead of

5.40 V. Alkaline cells are a bit lower, and I think that NiMH are even lower, thus closer to the design voltage. (I've wondered how much I could get away with without burning something out. Your tech friend may have included a series silicon diode to drop about 0.700V to make it a bit closer to design voltage.

For the height gauge, I would probably take a wall wart (those fat plugs which power a lot of things) and add a regulator to feed it the desired 5.40 V -- assuming that you do not have to use it somewhere where wall power was not available. A cord on a hand-held digital caliper is a bit awkward. But I have been planning to make a replacement holder to use two CR2032 cells (3.0 V each), which would retain the portability. And -- I would add a switch to allow me to set it down and pick it up a month later without needing to put in new cells. :-)

I loved the B&S calipers as well -- which is why I took so much time to learn about mine. There are features in the newer ones -- such as the Mitutoyo "absolute" which lets you switch back and forth between a preset value and a true zero without having to re-zero each time. But I really like the B&S ones. (I have two of them now, having picked one up for free at a metalworking yard sale, after explaining to the owner (who I knew/know) about the cells made of unobtanium. He offered it to me -- I did not ask him for it -- as he felt it better for it to go to someone who knew about the problems. My original one came from a hamfest, in a fitted mahogany case. The other is in some blonde wood instead. Otherwise, the cases are identical

And -- I have verified that it will work on 6.000 V, so I now need to figure out the overvoltage with two CR2032 cells instead. They are nominally 3V each, but in reality go a bit higher.

O.K. It looks like about 3.1V or a bit less -- under the extreme conditions of 60 degrees C. At 20 degrees C, it looks closer to

2.95V, so no problem. (This with a 190 uA for a discharge curve which goes out to a bit over 1100 hours and discharge down to about 2.0 V.

Enjoy, DoN.

Ed, if your still in need of the repair, I found a place in California that repairs the Digi-hite Plus. It's called Nyco systems. Their website is

thier phone number is 866-711-6797, they quoted me $270.00 for the repair. Look them up if you still need it repaired. Best of luck.