Retrofitting a 1986 Maho MH400E

Hello Bernd,

Cool machine. If you take the cover off the RH side (fwd looking aft) you should expose the Y Axis (normal naming - MAHO calls it Z) linear encoder. It will be interesting whether you machine has Heidenhain, or Phillips encoders. That info might also be in your manuals.

Mark

Hi Mark,

today I disassemble all the covers and sheet metal work, to clean everything and opened also the covers on the scales.
Im the less fortunate one, and got a machine with Philips 4820 heads.
As mentioned in the "Retrofit MAHO MH700c with Philips 432 control" on this forum, its very difficult to convert the signals like you did with the reused EXE from Heidenhain.
And the prices for spare pars for the Philips scales are catastrophic.

The next problem is, two Drive cards, the LM/RM and the RM/RM are broken, and I can't check the Scales without them (at least I'm not capable to do it), so i think I'm going to install new Scales with EIA422 output.

What I don't know is, if the Philips scales also have a Index for homing, ore if this Version of machines uses the first switch on the end of travel to do this, so I could also use them for homing.

Mark, as you are already very advanced with your retrofit, would you like to send me a copy of your schematics?
It would be a real big help, if I could use more or less the same I/O' numbers like you did, so if I'm in trouble (and for sure I will get to this point), I maybe can disturb you and ask you some questions.

I ordered a Mesa 6i25, 7i77 and a 7i84 for I/O expansions, so our circuits could look very similar.

Bernd

Bernd, I also have a MAHO with Philips Scales. I only powered it up once and it seemed to be OK. Do you have any documentation on the scales e.g. a pinout of the cables etc.? If you want to I could lend you my cards so that you can check your scales.
The scales always have an index, a simple switch would be far away from the desired accuracy. But some of the "newer" once have multiple indices so that at power up only a little axis movement is necessary to gather the actual position (as long as the machine remembers where it stopped before power of)

Christian

Hi Christian,

no, I only have the Maho Documentation, but reading again, I saw they describe the Scales with "periodisch-absolutes Wegmeßsystem" so this means it reads absolute values, on a length of 0,635mm. It uses a switch, near end of travel to select the right index for homing.

They describe also, that the signal from the reading head is converted on a special electronic device to a PWM-Signal, and go's then to the control, in there it's translated to Indexpulses, counting-pulses, and direction of travel.

Are you reusing the Philips-scales, and what hardware do you need to to this? As far as I understood, the equivalent hardware to Marks EXEmodul (if there is any?) sits directly on the LM-Drivecards?

It would be a good idea, to check the machine with your drive cards, if the Philips-scales are reusable on LinuxCNC..
Do you know where the battery is hidden? The Control lost all the machine constants, and if i'm going to rewrite them by hand -i dont have a punch tape reader- I don't want to loose them again after a restart.

Bernd

The only thing I know about the Phillips scales was what MuellerNick wrote about them when he converted his 700C. You might want to read his thread through google translate... (you will need to sign up to read it).
www.cncecke.de/forum/showthread.php?4346...ll-nach-Puchschlagen
I think he used a mix of MESA 5i20 und 7i33, but also rolled some of his own interface hardware (very impressive work)

Might be a good idea to collect all the MAHO x00C model info into it's own thread. So far this thread concentrates on the MHx00E's which all seem to be quite similar.

I finished cleaning my linear enoders, and got Y back installed again. It worked perfectly again, so I left it powered up and sure enough, after about 10 minutes it is acting up and dropping most steps again. I still didnt get around to scoping the signal, to see which channel or both is dropping out.
Mark

I confirmed that behaviour of the Y axis encoder yesterday. Started the machine, homed the axis, everything fine. Set up the HAL Osci, and there were already drop outs on the B channel. Played around with it a bit and B dropped out completely, along with CH A starting to have drop outs. Double checked the X encoder with the HAL Osci and it is still fine. When I went back to the Y axis, the signals were basically gone. Got a single step on Ch A, enough to trigger the oscilloscope, but that was it.

I looks like flat batteries
I was thinking about this later, and the 5V PSU to drive the 7i77 and encoders is only 4.7V measured at the EXE board. Heidenhain gives the tolerance of +/-5% on the 5V 120mA power requirement for the LS-403 encoder, so it should be in tolerance. The 5V PSU I added does have a voltage adjust, but I didn't think of it at the time. I'll tweak that up and test again.
Mark

It is a tiny incandescent bulb. 5V 0.6W. When I had the heads out for cleaning, I left them powered for quite a qhile, and there was no obvious dimming.

Hi to all, is this a good idea for spindle encoder on mahos?