Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley

Hi Caudlet,

That's what I thought! It just multiplies, and offers a choice of
common +5 or common 0V.

So I still need a way to divide the encoder quadrature signals!

Alan KM6VV


caudlet wrote:

>
> --- In CAD_CAM_EDM_DRO@yahoogroups.com, Alan Marconett <KM6VV@a...>
> wrote:
>
>>Hi Lance,
>>
>>You're saying that the pulse multiplier in the G340 both multiplies
>
> the
>
>>pulses from the CPU, AND divides the pulses coming in from the
>
> encoder?
>
>> I guess I missed that. I was under the impression that the
>
> little
>
>>board was only slightly different from the one in the G210, called
>
> a
>
>>G901. The notes for the G901 don't mention any divider for the
>
> encoder.
>
>>My servos are 300W Sanyo Denki "Super R". I haven't found any
>>additional information on them, other then what's on the nameplate.
>>Maybe I don't need any pulley reduction on them, they probably have
>>considerable torque!
>>
>>I was planning on trying them out on the original leadscrews (20
>
> TPI?)
>
>>on my RF-31 mill (ugh!), before I can get ballscrews. Probably
>
> means I
>
>>need a LOT of steps!
>>
>>8000 * 20 is 160,000 counts per inch.
>>60 IPM would need 160K steps/sec.
>>
>>Sounds like a Gecko G340 could handle that, but there were some
>
> comments
>
>>about positioning errors at high pulse rates (over 250K?). And I
>
> don't
>
>>think my old computers are up to that! If I do any belt reduction,
>
> I'm out.
>
>>Yeah, I want to play with some G340s!
>>
>>Alan KM6VV
>>
>>
>
> Alan nothing is divided. The board in the 210 and 340 muiltiplies
> the incomimg steps so your computer doesn't have to try and output
> more than it can (normally 25 to 45,000 pps). It works like this.
> send the multiplier card a pulse and depending in its setting it will
> send 2, 5 or 10 pulses to the normal motor drive circuit. One pulse
> in gets you multiple pulses out of the multiplier card. Since the
> upper limit of the normal Gecko circuit is 250,000 pps from ANY
> source (straight input or from the multiplier board) you could not,
> for example, stuff 40,000 pps in with the multiplier set at X10 and
> try to feed the base circuit 400,000pps second. The output of the
> encoder has four distinct transitions per 'line' so if you have 2000
> lines you get 8000 pulses per rev.
>
> This combination is not very well matched. Kinda like putting 30"
> tires on a shopping cart. You have way to high a line count on the
> encoders and the 20 TPI screws are an added limiting factor. The good
> news is you should not face any torque issues!
>
>