Re: [Emc-users] resolver/encoder converter
On Friday 08 February 2008, Jon Elson wrote: >Gene Heskett wrote: >> On Thursday 07 February 2008, Jon Elson wrote: >>>Gene Heskett wrote: I have about 6 feet of chip books (yeah, I know, today it would take 300 feet of shelves for half of them), Jon, but they don't include Analog Devices, and they are getting long in the tooth, sorta like me. :) I guess I'll have to go begging again. >>> >>>I don't use data books anymore. All the data sheets are on >>>line, and you are guaranteed to get the latest version, etc. >>>30 seconds with Google will get you to analog devices, and you >>>can look up on their site keywords of resolver and digital, and >>>you'll get a check list of the applicable models and the most >>>important differences between models. >>> >>>Jon >> >> I know that Jon, but old habits die only with difficulty it seems. >> Darnit. > >Well, I've been designing some new stuff at work using these >chips that are no bigger than a grain of rock salt (3 x 3 mm). >So, I've started using some stuff that is so new they are not in >ANY databook, only on-line. > >TI has gone insane, at last count there were 136 pages of TI >logic chips, at at least 100 part #'s per page, that is 13000+ >different parts! (Yes, some part numbers specify tape and reel >rather than cut tape, so it may be a little less. But, totally >insane to make so many oh-so-slightly different parts. I don't >understand. > >Jon All I can say is WOW & I don't understand it either! They must be trying to fit every socket on the planet, and that is not good business sense IMO. And that's just logic? What about analog & discretes? Whatever became of 'if you wanted cmos, call rca, if you wanted an op-amp, call Nat Semi, level translators call Mitel, power hexfets call IR" etc etc.. Note that's a statement even though it reads like a question. :) -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Since aerosols are forbidden, the police are using roll-on Mace! - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
Stuart Stevenson wrote: > Gentlemen, > I purchased an evaluation board from Analog Devices. I will try > it. It is the eval-ad2s1200cbz. It is fully populated with a BUNCH of > stuff. Randy and I will work with it. The tech support said it will > work. Randy read the propaganda and said it will work. I will know > Monday evening. Don't be so sure. Resolvers are fairly tricky, and the signals coming back are typically pretty small, so you have to watch shielding and grounding carefully. I hope it is a simple drop in, but if not, you should be able to get it going without anything extreme. I'm kind of interested in these, so I would like to hear how it goes. (If you want a minimal commercial board made for this, I'd be interested in working on it. Just resolver in, maybe a couple jumpers to set reference frequency, and some jumpers for the resolution.) Jon - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
Gene Heskett wrote: > On Thursday 07 February 2008, Jon Elson wrote: > >>Gene Heskett wrote: >> >>>I have about 6 feet of chip books (yeah, I know, today it would take 300 >>>feet of shelves for half of them), Jon, but they don't include Analog >>>Devices, and they are getting long in the tooth, sorta like me. :) I >>>guess I'll have to go begging again. >> >>I don't use data books anymore. All the data sheets are on >>line, and you are guaranteed to get the latest version, etc. >>30 seconds with Google will get you to analog devices, and you >>can look up on their site keywords of resolver and digital, and >>you'll get a check list of the applicable models and the most >>important differences between models. >> >>Jon >> > > I know that Jon, but old habits die only with difficulty it seems. Darnit. > Well, I've been designing some new stuff at work using these chips that are no bigger than a grain of rock salt (3 x 3 mm). So, I've started using some stuff that is so new they are not in ANY databook, only on-line. TI has gone insane, at last count there were 136 pages of TI logic chips, at at least 100 part #'s per page, that is 13000+ different parts! (Yes, some part numbers specify tape and reel rather than cut tape, so it may be a little less. But, totally insane to make so many oh-so-slightly different parts. I don't understand. Jon - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
Gentlemen, I purchased an evaluation board from Analog Devices. I will try it. It is the eval-ad2s1200cbz. It is fully populated with a BUNCH of stuff. Randy and I will work with it. The tech support said it will work. Randy read the propaganda and said it will work. I will know Monday evening. thanks Stuart - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
On Thursday 07 February 2008, Kirk Wallace wrote: >On Thu, 2008-02-07 at 12:41 -0600, Jon Elson wrote: >> Gene Heskett wrote: >> > I have about 6 feet of chip books (yeah, I know, today it would take 300 >> > feet of shelves for half of them), Jon, but they don't include Analog >> > Devices, and they are getting long in the tooth, sorta like me. :) I >> > guess I'll have to go begging again. >> >> I don't use data books anymore. All the data sheets are on >> line, and you are guaranteed to get the latest version, etc. >> 30 seconds with Google will get you to analog devices, and you >> can look up on their site keywords of resolver and digital, and >> you'll get a check list of the applicable models and the most >> important differences between models. >> >> Jon > >Many years ago when I was a drafter, I would go over to the engineer's >area and go through the trash to find data books that I didn't already >have. I lugged those books around for far too long, though I kept a >couple for entertainment. Yup, some of the comments in Nat Semi books are like MasterCard, priceless. :) Need a fast analog buffer? They had 2 at the time, "fast", and "damned fast". Slew rate was 6000v/u-sec. Heck, we've got $1.25 op-amps faster that that now. >The problem now with is that there are far too many choices. It's way >faster to find what you need, but much longer to decide which one. I >like the good ol days when you had to live with what you couldn't have. Yeah, that too. I recall I needed a cmos analog multiplexer once, to switch video sources in a character generator, and the only way I could get the speed I wanted was to run 4000 family cmos stuff on the full supply available, in this case 28 volts. That's 15 volt rated stuff. But I could wrap a colored border around the character in a character generator when I got done. The 4028 warmed up about 10 degrees F, but everything ran cool and that way with zero failures for quite a few years after I built it. At 28 volts, it was the fastest cmos stuff on the planet. :-) -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Words have a longer life than deeds. -- Pindar - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
On Thursday 07 February 2008, Jon Elson wrote: >Gene Heskett wrote: >> I have about 6 feet of chip books (yeah, I know, today it would take 300 >> feet of shelves for half of them), Jon, but they don't include Analog >> Devices, and they are getting long in the tooth, sorta like me. :) I >> guess I'll have to go begging again. > >I don't use data books anymore. All the data sheets are on >line, and you are guaranteed to get the latest version, etc. >30 seconds with Google will get you to analog devices, and you >can look up on their site keywords of resolver and digital, and >you'll get a check list of the applicable models and the most >important differences between models. > >Jon > I know that Jon, but old habits die only with difficulty it seems. Darnit. -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Words have a longer life than deeds. -- Pindar - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
On Thu, 2008-02-07 at 12:41 -0600, Jon Elson wrote: > Gene Heskett wrote: > > > I have about 6 feet of chip books (yeah, I know, today it would take 300 > > feet > > of shelves for half of them), Jon, but they don't include Analog Devices, > > and > > they are getting long in the tooth, sorta like me. :) I guess I'll have to > > go begging again. > > > > I don't use data books anymore. All the data sheets are on > line, and you are guaranteed to get the latest version, etc. > 30 seconds with Google will get you to analog devices, and you > can look up on their site keywords of resolver and digital, and > you'll get a check list of the applicable models and the most > important differences between models. > > Jon Many years ago when I was a drafter, I would go over to the engineer's area and go through the trash to find data books that I didn't already have. I lugged those books around for far too long, though I kept a couple for entertainment. The problem now with is that there are far too many choices. It's way faster to find what you need, but much longer to decide which one. I like the good ol days when you had to live with what you couldn't have. -- Kirk Wallace (California, USA http://www.wallacecompany.com/machine_shop/ Hardinge HNC lathe, Bridgeport mill conversion, doing XY now, Zubal lathe conversion pending) - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
Gene Heskett wrote: > I have about 6 feet of chip books (yeah, I know, today it would take 300 feet > of shelves for half of them), Jon, but they don't include Analog Devices, and > they are getting long in the tooth, sorta like me. :) I guess I'll have to > go begging again. > I don't use data books anymore. All the data sheets are on line, and you are guaranteed to get the latest version, etc. 30 seconds with Google will get you to analog devices, and you can look up on their site keywords of resolver and digital, and you'll get a check list of the applicable models and the most important differences between models. Jon - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
Stuart Stevenson wrote: > The 50 ms came from a conversation with Matt in tech support at AMCI. > I wrote it down as he was saying it. He said the converter was > developed in concert with Rockwell specifically to replace the servo > amp I have with a different servo amp. The converter had too much > delay to work in the application it was being developed for. I will > probably try to return them if they will not work. Something is quite odd, here. I can't believe any modern servo amp could tolerate a 50 ms lag. it would require the servo amp to have a bandwidth somewhere around 5 Hz. Even my 1978 Allen-Bradley control did better than that. Ah, yes, I suppose it DID have "too much delay" to work. You probably won't get this info, but it is at least a parting shot across the bow you can throw at them, ask how many of these things they've sold. If the answer is non-zero, ask how many are installed in actual servo control loops. I would make a strong wager that number is precisely zero. The only possibility I can imagine is the guy who designed this thing had never been told it was going to be used in a servo control loop, and never thought about the delay until the project was done. Then, somebody said "Oh, what is the delay from resolver movement to output", and there were a bunch of "oops-es!" Jon - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
On Wednesday 06 February 2008, Jon Elson wrote: >Gene Heskett wrote: >> On Wednesday 06 February 2008, Alex Joni wrote: These introduce a 50 ms delay in the feedback signal. Is this acceptable? I think, probably not. My question is this. Do I need to install encoders on the motors? Questions, thoughts, comments welcome. >>> >>>50msec is waay too much, but reading the >>>http://www.amci.com/pdfs/accessories/RD141S_rev_1.0.pdf >>>I can't find it... are you sure it wasn't 50ns ? >> >> I doubt if its that fast. With a 5khz carrier, I'd assume it takes a full >> cycle to come up with an angle and then convert, and the angle will take >> .2 milliseconds if its done in a single full cycle of the carrier. >> >> Or do I not understand how these work? > >I've studied the Analog Devices units. They have a tracking PLL >that attempts to track the velocity of the resolver. So, it >doesn't quite do a complete conversion from scratch every cycle >(or half-cycle) of the sensor excitation. It is more like it >generates an estimate of position from the PLL and then refines >the error. >Also, using the AD chip, you can set the frequency to whatever >you want. My guess is that AMTI clunker has a PIC micro with >A/D in it, and not much else, and costs them $15 to make. I >didn't see what Stuart had to pay for them. But, if that 50 ms >lag is correct, they are of VERY limited usefulness. > >Jon I have about 6 feet of chip books (yeah, I know, today it would take 300 feet of shelves for half of them), Jon, but they don't include Analog Devices, and they are getting long in the tooth, sorta like me. :) I guess I'll have to go begging again. -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) "You can't teach seven foot." -- Frank Layton, Utah Jazz basketball coach, when asked why he had recruited a seven-foot tall auto mechanic - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
> Message: 7 > Date: Wed, 6 Feb 2008 18:05:51 +0200 > From: "Alex Joni" <[EMAIL PROTECTED]> > Subject: Re: [Emc-users] resolver/encoder converter > To: "Enhanced Machine Controller \(EMC\)" > > Message-ID: <[EMAIL PROTECTED]> > Content-Type: text/plain; format=flowed; charset="iso-8859-1"; > reply-type=original > > > These introduce a 50 ms delay in > > the feedback signal. Is this acceptable? I think, probably not. > >My question is this. Do I need to install encoders on the motors? > >Questions, thoughts, comments welcome. > > 50msec is waay too much, but reading the > http://www.amci.com/pdfs/accessories/RD141S_rev_1.0.pdf > I can't find it... are you sure it wasn't 50ns ? The 50 ms came from a conversation with Matt in tech support at AMCI. I wrote it down as he was saying it. He said the converter was developed in concert with Rockwell specifically to replace the servo amp I have with a different servo amp. The converter had too much delay to work in the application it was being developed for. I will probably try to return them if they will not work. > > Regards, > Alex > > > > > -------------- > > Message: 8 > Date: Wed, 6 Feb 2008 08:22:52 -0800 > From: Dave Engvall <[EMAIL PROTECTED]> > Subject: Re: [Emc-users] resolver/encoder converter > To: "Enhanced Machine Controller (EMC)" > > Message-ID: <[EMAIL PROTECTED]> > Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed > > The Analog Devices 2S2105 resolver to digital chip with encoder > emulation shows a max latency of 125 us. > It is supposed to track at 1000 r/s. These are about $20 chips and > not the top of the line. > HTH > I will definitely check this out. thanks Stuart - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
Gene Heskett wrote: > On Wednesday 06 February 2008, Alex Joni wrote: > >>>These introduce a 50 ms delay in >>>the feedback signal. Is this acceptable? I think, probably not. >>> My question is this. Do I need to install encoders on the motors? >>> Questions, thoughts, comments welcome. >> >>50msec is waay too much, but reading the >>http://www.amci.com/pdfs/accessories/RD141S_rev_1.0.pdf >>I can't find it... are you sure it wasn't 50ns ? >> > > I doubt if its that fast. With a 5khz carrier, I'd assume it takes a full > cycle to come up with an angle and then convert, and the angle will take .2 > milliseconds if its done in a single full cycle of the carrier. > > Or do I not understand how these work? I've studied the Analog Devices units. They have a tracking PLL that attempts to track the velocity of the resolver. So, it doesn't quite do a complete conversion from scratch every cycle (or half-cycle) of the sensor excitation. It is more like it generates an estimate of position from the PLL and then refines the error. Also, using the AD chip, you can set the frequency to whatever you want. My guess is that AMTI clunker has a PIC micro with A/D in it, and not much else, and costs them $15 to make. I didn't see what Stuart had to pay for them. But, if that 50 ms lag is correct, they are of VERY limited usefulness. Jon - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
Stuart Stevenson wrote: > The opportunity is this. To get quadrature feedback into EMC2 from > the motors I will need a resolver to encoder converter or I will need > to install encoders on each motor on the X, Y and Z axes. I purchased > three converters, RD141S from AMCI. These introduce a 50 ms delay in > the feedback signal. Is this acceptable? I think, probably not. 50 ms WHAAAT? These must be made for DRO use only, not CNC. 50 ms is 50 default servo sample times, so no way would you be able to tune EMC2's servo response to work with this. > My question is this. Do I need to install encoders on the motors? My understanding of the Analog Devices resolver-digital chips is they use a phase locked loop to track the resolver position, and that will produce quadrature output that has very little lag from the real shaft position. So, if you can find a converter box that uses the AD chips it should get rid of most of the lag. These chips used to be insanely expensive, but they have come way down since about 2000. Their latest chip is an all-in-one solution, with no need for a CPU at all, and it produces ABZ right off the chip. Jon - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
On Wednesday 06 February 2008, Alex Joni wrote: >> These introduce a 50 ms delay in >> the feedback signal. Is this acceptable? I think, probably not. >>My question is this. Do I need to install encoders on the motors? >>Questions, thoughts, comments welcome. > >50msec is waay too much, but reading the >http://www.amci.com/pdfs/accessories/RD141S_rev_1.0.pdf >I can't find it... are you sure it wasn't 50ns ? > I doubt if its that fast. With a 5khz carrier, I'd assume it takes a full cycle to come up with an angle and then convert, and the angle will take .2 milliseconds if its done in a single full cycle of the carrier. Or do I not understand how these work? >Regards, >Alex > > >- >This SF.net email is sponsored by: Microsoft >Defy all challenges. Microsoft(R) Visual Studio 2008. >http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ >___ >Emc-users mailing list >Emc-users@lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/emc-users -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Postmen never die, they just lose their zip. - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
The Analog Devices 2S2105 resolver to digital chip with encoder emulation shows a max latency of 125 us. It is supposed to track at 1000 r/s. These are about $20 chips and not the top of the line. HTH Dave On Feb 6, 2008, at 8:05 AM, Alex Joni wrote: >> These introduce a 50 ms delay in >> the feedback signal. Is this acceptable? I think, probably not. >>My question is this. Do I need to install encoders on the motors? >>Questions, thoughts, comments welcome. > > 50msec is waay too much, but reading the > http://www.amci.com/pdfs/accessories/RD141S_rev_1.0.pdf > I can't find it... are you sure it wasn't 50ns ? > > Regards, > Alex > > > -- > --- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ > ___ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] resolver/encoder converter
> These introduce a 50 ms delay in > the feedback signal. Is this acceptable? I think, probably not. >My question is this. Do I need to install encoders on the motors? >Questions, thoughts, comments welcome. 50msec is waay too much, but reading the http://www.amci.com/pdfs/accessories/RD141S_rev_1.0.pdf I can't find it... are you sure it wasn't 50ns ? Regards, Alex - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
