Retrofitting a 1986 Maho MH400E

Thanks Mark!

Your thoughts are similar to mine in regards to the design and replaceability of the single-axis boxes. My plan, if the fault is confirmed to be EXE-related, is exactly as you say, swapping one of the channels with a compatible module while leaving the rest alone. In that case, the remaining channel on the PCB is simply unplugged and doesn't do anything, AFAIK.

If that happens, I'll see if there are manuals on the Heidenhain website with detailed info about which serial numbers do what. There's bound to be some kind of coding scheme that can tell you which modules output 1 micron resolution signals and which ones do 5. There doesn't seem to be a price difference between the listings of 60x EXEs on ebay. I wouldn't be surprised if the people selling them don't really know what they do. The more expensive ones are simply new old stock, while the cheaper ones are used. The former might be worth it, but the price difference is about a 100%, so maybe they aren't

Hi Marcin,

I have collected a fair bit of Heidenhain data. Like the attached.

Checking this again, especially the technical description on page 2, you can see that the interpolation is hidden in the part numbers, not in the model numbers. Looks like any model could be either 5µm or 1µm, so you really need to research the part number advertised before buying.

My original 3 Axis EXE is only decoding my X-Axis. The other two channels, one is dead, the other flaky. If I knew I could get two more 1µm EXE's for€50 each, I'd remove the 3 axis EXE and the 5µm 602/1-F, but that is just cosmetics.
Mark

When I read the documentation, there comes upon me an inexorable sense that the source of my problems is a hack job in the electric cabinet. In those moments, it's best to dive into the comfortable, if misguided, notion that consumerism holds the actual answer to the issue at hand and leave things be until such a time that sufficient strong will and free time are mustered to investigate matters beyond cursory impressions and gut feelings.

Well you understand my "trouble shooting" methods master Yoda!

Hi There,

check pages 18/19 of Mark's manual. Switches S1/S2 are used to change between 1-fold (=5µm) and 5-fold interpolation (=1µm). I wonder whether they used the same PCB for both versions? If your 602/1-F has these switches it might be worth a trial.

Oh, good pick up. I'll need to check out if I can switch the second EXE to 1µm output. That would be easy.
Mark

I've got to check my library but I am pretty sure these will do:
www.ebay.de/itm/HEIDENHAIN-Impulsgeber-EXE-602-E/183839235112

Or you can get the units I got, EXE 610 C. They have 10 times the resolution (of the 602), and with the standart scales with 20 um repetition give a resolution of 0,1 um!

But consult the documentation for the exact model numbers to make sure you get the ones with the correct resolution. There are many alternatives out there. Also ones where you can sett the resolution to many alternatives.

On my mill this gives a smoother movement on low speeds. I definately know that I won't be able to achieve 10 times the accuracy by doing this. But resolution on slow movements is better, and velocity is calculated more precise.
For more details on my findings have a look in the thread over at PM. I wrote more details about my findings on this there, and don't feel like repeating that (and it's a long time ago so I don't even remember the details that well anymore).
I just had a quick look, and you get these just as cheaply as the 602. Even though someone thinks they can get $500 or more for them..... they go lower than $50, and there is no reason to pay a premium for them. Either they work or they don't. It's not an item that wear out....

Anders

That is a great input Anders. Thanks. Given that a MESA 7i77 and 5i25 has no speed problem processing the higher resolution, this would be the way to go. If you want to get really crazy, the 660B series can be configured for 400x interpolation.

My 1µm 602D/5-F has the internal switches for selecting 1µm or 5µm resolution.






... whereas the 602/1-F is obviously older from the size of the IC's and has a row of soldered jumpers.




I would bet that a different configuration of jumpers would provide the higher resolution. There is provision for 5 jumpers, which matches with the later series 5 switches, but the change need to modify it from 5µm to 1µm isn't obvious, given that the jumpers from each end are currently one connected and one open, this does not map to the switch configurations of the later EXE's.
Mark

Don't want to interrupt the conversation.
Just wanted to post pictures of some cosmetics I did today.

Before:


After:


First chips I made with this machine: