stepper drive encoder support

[patrick71100]

Any Of You Sharp Guys Know Of A Stepper Drive That Can Use Encoder For Position Verification I Have Grex But Stepper With Encoder Is Not Supported Yet,or Just Help With Drive To Use For Big Slo-sync 1700 Oz Holding Stepper For 4 Axis 5c Collet Rotory I
Plan To Use Gates Gt2 Timing Belt 1:3 Ratio To Elimanate Backlash
Compared To Worn Worm Drive,thanks Shawn

[NC Cams]

Go to Rogers Machine.

They have a stepper/encoder interface board.

[tekno]

would you have a link?

Originally Posted by NC Cams Go to Rogers Machine. They have a stepper/encoder interface board.

[NC Cams]

Rogers Machine has been mentioned MANY times on this M/B.

Do a Rogers Machine with the site's search engine.

[ger21]

www.rogersmachine.net

But, it won't work with a G100, because afaik, you can't use a parallel port when using the G100.

[Davidh]

Could you share why encoders are necessary, My stepper driven machine can run for 2 hours and always 'comes home' by itself. Is it case of belt and braces? Or are you having problems with steps?

[eidos]

this board is a FAKE !!
mach 2.3.4 not support "closed loop" , but suport encoders only in jog manual mode.
hi

Originally Posted by ger21 www.rogersmachine.net But, it won't work with a G100, because afaik, you can't use a parallel port when using the G100.

[ger21]

Originally Posted by eidos this board is a FAKE !! mach 2.3.4 not support "closed loop" , but suport encoders only in jog manual mode. hi

The Rogers board comes with a macro that runs in Mach3 and compares the encoder position to the commanded position. If the deviation is greater than a user specified amount, the macro will pause Mach3 to prevent it from ruining your part.

The board isn't really necessary, it just makes it easier to connect the encoders. The macro is what does the work, and could be written by any user with a little programming experience. All it's doing is comparing numbers.

[NC Cams]

Perhaps its time to re-review some of the basics:

http://www.cnczone.com/forums/showthread.php?t=17419

BOTH steppers and servos have their place in CNC. Each has merits and detractions.

Steppers with encoder feedback are, IMO, psuedo servos at best. For some applications, the're quite adequate and remarkably well behaved.

However, for some applications, as in OD and or cam grinding, steppers leave steps - period. You can literally 'see' them and in some cases "hear them" as a roller rolls over the surface (sounds like baseball card in bicycle spoke). If you want a beatiful glass smooth CNC'd surface, you're talking high end servos - period.

I've yet to see a low end, turn key, don't have to write code, CNC retrofit with servos that incorporates full feedback at a DIY afffordable price offered yet - affordable as in the price range of Mach.

Ajax claims $2800 entry fee - bull puck. See prior lathe retrofit thread for what ultimately happens. Figure more like $8-$10k and less than perfect lathe performance if you believe the users who reported their issues in the thread.

The only stepper encoder of any note that is offered is the Rogers Machine piece. As posts 7 and 8 show, it is not the "nirvana" that some people seek.

The stepper vs servo debate lumbers on....

[Torchhead]

The main challenge in using the encoders as feedback with Gecko's (and several other drive types) is that the encoder signals (A + B) are referenced to the Gecko internal buss. Since the Gecko has opto inputs the PC and logic level ground (because of noise prevention) may not have the same ground as the Motor DC, and tying the encoders directly back to the logic input can be destructive to one or the other system. The Rogers card isolates the signals and allows you to breakout the encoder signals without having to worry about frying electronics (been there, done that). While the Macro is the heart of the software interface it's important not to ignore the electronic interface and to protect the involved circuits. The optos involved need to have enough bandwidth to support the encoder frequency. Since the gecko has limited +5 for encoders, an external source of higher current +5 sometimes is required. Once again it has to be ground referenced to the Gecko negative buss. If you opto isolate the encoders off the Gecko buss then the encoders can be run off the PC commom +5 and the current source problem goes away. These are approaches that the designer of the drives recommends.

No, Mach3 does not support software feedback loop of the drives. Properly designed stepper systems used inside their specificiations do not lose steps. So the only time you would need the system to apply correction is if you had a condition that was outside the capabilies of the drives to begin with....so what would you have it do under those conditions? Apply power it doesn't have? Try greater acceleration than it can do?. Lost steps are the result of some condition that the drive already cannot correct for. If you have a system that loses steps you need to find and fix the problem. The Rogers card as pointed out just halts, keeps from ruining a part and alerts you that steps have been lost so you can find the problem.

[ger21]

Originally Posted by Torchhead The main challenge in using the encoders as feedback with Gecko's (and several other drive types) is that the encoder signals (A + B) are referenced to the Gecko internal buss. Since the Gecko has opto inputs the PC and logic level ground (because of noise prevention) may not have the same ground as the Motor DC, and tying the encoders directly back to the logic input can be destructive to one or the other system. The Rogers card isolates the signals and allows you to breakout the encoder signals without having to worry about frying electronics (been there, done that). While the Macro is the heart of the software interface it's important not to ignore the electronic interface and to protect the involved circuits. The optos involved need to have enough bandwidth to support the encoder frequency. Since the gecko has limited +5 for encoders, an external source of higher current +5 sometimes is required. Once again it has to be ground referenced to the Gecko negative buss. If you opto isolate the encoders off the Gecko buss then the encoders can be run off the PC commom +5 and the current source problem goes away. These are approaches that the designer of the drives recommends.

The original post was talking about steppers, so this doesn't apply, since Gecko stepper drives don't have any encoder connections.

[eidos]

hi all
external macro to Mach 3?
who is developing it? the same programmer of mach 3?
or furnished together with the suitable card?

if so pits, would be a good thing, for a finally available closed loop also on a software that works on parallel port, it would be enough to directly integrate the macro in mach 3, rather than to have an external software that can always slow down mach while working.....

hi

Originally Posted by ger21 The Rogers board comes with a macro that runs in Mach3 and compares the encoder position to the commanded position. If the deviation is greater than a user specified amount, the macro will pause Mach3 to prevent it from ruining your part. The board isn't really necessary, it just makes it easier to connect the encoders. The macro is what does the work, and could be written by any user with a little programming experience. All it's doing is comparing numbers.