Those are cute encoders and indeed the right size. Probably spendy. I chose the direct route since I’ve already used encoders on my small machine. The antibacklash system on the resolver mount assumes the case is eccentric to the shaft so one removes the backlash by rotating the encoder. These resolvers were gear driven and by going to the encoders I get a direct coupling to the servo shaft. I like simplicity! :-)
I understand the Vigilant or something like that makes black boxes that are resolver to counts. Will even provide the right connectors. The price I’ve heard is about $300/channel.
The Mazak has a pulse code handwheel: sure would be nice to integrate that into emc, as if I need more on my plate. Just getting the basics running is going to be a challenge.
Good luck with your conversion.
Dave
Dave,
I recently noticed that someone sells an encoder that intended as a resolver replacement. I have not checked price, but I thought I might use some (someday) for a LeBlond lathe retrofit that is sitting on the back burner right now. It uses a total of 5 resolvers and/or synchros. This would allow the same physical mounting as the resolvers, but with the desired encoder output.
No mechanical changes would be a good thing in my case, due to the mounting arraignment of at least two of them. They are mounted to hand wheels through a gear train that also is motor driven when using the hand wheels in "power feed" mode.
I had planned to use resolver to encoder conversion ICs (Analog Devices makes some - ADS90 or something I think), or a commercial conversion board (Computer Conversions, I think). Neither of these choices seemed as direct a solution as physical replacement. I guess if the cost and performance is similar (if not better), less components is the preferred option.
They also make a handheld pulse generator with selector switches for axis select and scale factor, and other options. I just bought one for a mill retrofit, but have not used it yet.
Here's a link...
http://www.cui.com/encoders/encd_tools.htm
Regards,
Steve
Dave Engvall wrote:
Hi Paul;
Thanks for the guidance. Will give it try sometime down the road. Am up to
my ears in alligators on the Mazak. It should be fun when I get it running
though. All of it takes time. I just got one of the resolvers swapped out
for an encoder. Only two more to go. Not much room to play with; I'm going
to try a modular encoder for the next one.
If someone is going to do a PCI replacement for the Dan's board they might
socket it for a 4 channel DAC: then someone has the option of populating it
later on.
Dave
On 12/19/02 12:38 PM, "Paul" <paul.corner-at-tesco.net> wrote:
>
>
> Hi Dave
>
> Take the extstgmot.c source code, and using it as an example, add extDAC**
> functions to extsmdromot.c - Include the extAio** functions as well, but
> these can just be stubs that "return 0;".
> In the emcmot/Makefile, I'd suggest using stgmod.o for this as it is the
> first one in the list - Replace the reference to stg.o with dro.o and do a
> "make PLAT=$RTPLAT".
>
> This will give you the basic analog outputs and encoder inputs and digital IO
> via the parallel port. The parport base address will be hard coded to 278 as
> defined in parport.h.
>
> Some of this is documented in the source notes in emc/documents, and also in
> a pdf doc to be found at http://www.linuxcnc.org/BDI/index.html
>
>
> Regards, Paul.
>
>
>
> On Thursday 19 Dec 2002 5:24 pm, Dave Engvall wrote:
>> The
>> code for the D/A is only a few lines but I didn't know enough to link it to
>> emc.
>