Re: [CAD_CAM_EDM_DRO] Re: Major encoder trouble, need advice.

Henrik Olsson wrote:

> Hello,
> Yes, although I've looked at the signals I probably won't be able to see
> and/or catch the exact moment where it goes wrong. The scope is didgital but
> honestly I don't know my way around it enough to set it up for this kind of
> one-shot events.... :-(
>
>

Most digital scopes don't have the cross-channel triggering qualifiers
like a logic analyzer to do this. One thing you COULD try to find in
their dizzying array of "features" is either a runt pulse trigger or
minimum pulse width. Most likely, the problem is one of those. A pulse
which only goes half way from one logic rail to the other, then goes
back, or a very narrow pulse. Most digital scopes should have a feature
to trigger on those.

> These motors are old but top quality German manufactured servo motors by
> Indramat, I don't want to know the price of a new one... I'm tired of
> messing with this so I've fired of an E-mail to Renco about their RM21
> encoder which is available with differential line driver and with 625lines
> which I want. Hopefully the price isn't too bad and they'll sell me three
> but we'll see. I've expressed my concern about noice so I'll have to wait
> and see what their response is.
>
>

I think you'll find their encoders reasonable, and you should NOT have
any trouble with them.
You will have to choose from a list of options for the right connector,
index or none, active pullup or passive, shaft or bore size and so on.

> The 26C32's will be here early next week, I'll give that a go but honestly
> don't think it will make any difference - you never know though.
>
>

I'd be darn surprised if it changes anything. I suspect that when the
power transistors in the UHU switch RIGHT when the encoder is making a
transition, then the combined power/ground noise right on top of the
current pulse to charge up the cable exceeds the local supply
"robustness" (could be at the diff driver itself, or more likely in the
photo sensor amp) and it produces a glitch. A robust drive would detect
a momentary invalid transition and wait for it to be corrected, but in
time-critical software, there is only so much you can do before the next
interrupt.

Jon