Home |
Search |
Today's Posts |
|
Metalworking (rec.crafts.metalworking) Discuss various aspects of working with metal, such as machining, welding, metal joining, screwing, casting, hardening/tempering, blacksmithing/forging, spinning and hammer work, sheet metal work. |
Reply |
|
LinkBack | Thread Tools | Display Modes |
#1
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
The servos on my mill have sinusoidal encoders which are not
compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i |
#2
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. |
#3
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
On 2010-06-04, Pete C. wrote:
Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i |
#4
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i ..0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. |
#5
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
On 2010-06-04, Pete C. wrote:
Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i .0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. that sounds bad! i |
#6
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i .0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. that sounds bad! i Bingo! That would be why the "big boys" use 2,000 CPR or better encoders. What is the CPR on your current encoders? |
#7
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
I've had good luck with these:
http://usdigital.com/products/encode...rotary/kit/e6/ I'd get a higher count per inch encoder. get one that gives you a very even count per inch movement number, like 20,000 or 50,000 or even 100,000. remember you get four counts per encoder pulse. get the differential option. use their wire and connectors. Don't be a cheap skate and use the PC 5 volt supply. get one just for the encoders. I usually get a multiple voltage out DC power supply. Seems like I always need 24, 10, 5 I don't know if EMC does analog (10V) or opto isolated digital I/O (24V) Karl |
#8
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
On 2010-06-04, Pete C. wrote:
Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i .0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. that sounds bad! i Bingo! That would be why the "big boys" use 2,000 CPR or better encoders. I am confused. Your accuracy number for my example was 0.02". If you increase the count by 3 times (as in your mention of 2,000 CPR) then the accuracy only improves to 0.006", also an unacceptable number. Something is not right. What is the CPR on your current encoders? |
#9
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
"Pete C." wrote: Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i .0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. that sounds bad! i Bingo! That would be why the "big boys" use 2,000 CPR or better encoders. What is the CPR on your current encoders? For a home machine, I would want accuracy at least to 0.001", so the normal servo error window would need to be ~0.0005" or so to be acceptable. The big commercial machines hold much tighter accuracies, to the extent of using liquid cooled temperature stabilized ballscrews, double servos with glass scales on the machine, etc. |
#10
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: On 2010-06-04, Pete C. wrote: Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i .0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. that sounds bad! i Bingo! That would be why the "big boys" use 2,000 CPR or better encoders. I am confused. Your accuracy number for my example was 0.02". If you increase the count by 3 times (as in your mention of 2,000 CPR) then the accuracy only improves to 0.006", also an unacceptable number. Something is not right. What is the CPR on your current encoders? I think the issue is encoder "lines" vs. "counts". The encoders on the machines I worked on were 2,000 line encoders, which as Karl noted could provide 8,000 counts per revolution. The US digital site uses counts, so presumably the 720 CPR encoder is 180 line? |
#11
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
I think the issue is encoder "lines" vs. "counts". The encoders on the machines I worked on were 2,000 line encoders, which as Karl noted could provide 8,000 counts per revolution. The US digital site uses counts, so presumably the 720 CPR encoder is 180 line? The terminology on this is terrible. Multiply the USdigital number by four to get counts that your control will see. karl |
#12
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
On 2010-06-04, Karl Townsend wrote:
I've had good luck with these: http://usdigital.com/products/encode...rotary/kit/e6/ I'd get a higher count per inch encoder. get one that gives you a very even count per inch movement number, like 20,000 or 50,000 or even 100,000. remember you get four counts per encoder pulse. OK, great. Just what I need. I printed out the datasheet and will measure mounting dimensions to order them. get the differential option. use their wire and connectors. OK Don't be a cheap skate and use the PC 5 volt supply. get one just for the encoders. I usually get a multiple voltage out DC power supply. Seems like I always need 24, 10, 5 I don't know if EMC does analog (10V) or opto isolated digital I/O (24V) I will check what Jon's PPMC does, but I think that he takes 5v differential signals. Makes perfect sense to me. Thanks Karl. i |
#13
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Karl Townsend wrote: I think the issue is encoder "lines" vs. "counts". The encoders on the machines I worked on were 2,000 line encoders, which as Karl noted could provide 8,000 counts per revolution. The US digital site uses counts, so presumably the 720 CPR encoder is 180 line? The terminology on this is terrible. Multiply the USdigital number by four to get counts that your control will see. karl Lines = Number of physical lines on the encoder disk Counts = Lines * 4 (leading and trailing edges of the two quadrature channel signals) |
#14
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
"Ignoramus21167" wrote in message ... On 2010-06-04, Karl Townsend wrote: I've had good luck with these: http://usdigital.com/products/encode...rotary/kit/e6/ I'd get a higher count per inch encoder. get one that gives you a very even count per inch movement number, like 20,000 or 50,000 or even 100,000. remember you get four counts per encoder pulse. OK, great. Just what I need. I printed out the datasheet and will measure mounting dimensions to order them. get the differential option. use their wire and connectors. OK Don't be a cheap skate and use the PC 5 volt supply. get one just for the encoders. I usually get a multiple voltage out DC power supply. Seems like I always need 24, 10, 5 I don't know if EMC does analog (10V) or opto isolated digital I/O (24V) I will check what Jon's PPMC does, but I think that he takes 5v differential signals. Makes perfect sense to me. Thanks Karl. i Did you notice in the reading to put a resistor and capacitor on the lines? be sure to do this. Unless you like weird noise issues. Karl |
#15
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
On 2010-06-04, Karl Townsend wrote:
"Ignoramus21167" wrote in message ... On 2010-06-04, Karl Townsend wrote: I've had good luck with these: http://usdigital.com/products/encode...rotary/kit/e6/ I'd get a higher count per inch encoder. get one that gives you a very even count per inch movement number, like 20,000 or 50,000 or even 100,000. remember you get four counts per encoder pulse. OK, great. Just what I need. I printed out the datasheet and will measure mounting dimensions to order them. get the differential option. use their wire and connectors. OK Don't be a cheap skate and use the PC 5 volt supply. get one just for the encoders. I usually get a multiple voltage out DC power supply. Seems like I always need 24, 10, 5 I don't know if EMC does analog (10V) or opto isolated digital I/O (24V) I will check what Jon's PPMC does, but I think that he takes 5v differential signals. Makes perfect sense to me. Thanks Karl. i Did you notice in the reading to put a resistor and capacitor on the lines? be sure to do this. Unless you like weird noise issues. Yes, I did see that, and I will do so. Should be easy with Jon's breakout. I hope that my servo motors' mounting holes match some available options options for that encoder, that way it will be all neat and easy. i |
#16
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
I hope that my servo motors' mounting holes match some available options options for that encoder, that way it will be all neat and easy. Na, neat and easy, and refitting old machines are mutually exclusive. Any algebra expert should know this. Karl |
#17
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
On 2010-06-04, Karl Townsend wrote:
I hope that my servo motors' mounting holes match some available options options for that encoder, that way it will be all neat and easy. Na, neat and easy, and refitting old machines are mutually exclusive. Any algebra expert should know this. I will take a caliper to it and I will know, hopefully tomorrow or so. i |
#18
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
On 06/04/2010 10:04 AM, Pete C. wrote:
Ignoramus21167 wrote: On 2010-06-04, Pete wrote: Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i .0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. The normal rule of thumb in servo design is that the encoder error should be no more than 1/10th the desired total system error. Your 128 count rule would mean that your 1/8" pitch ball screw would give an error of 128 * 0.125" / 2 / 2000 = .004". Is that really what you meant? Four mils error? 'course, by my "ten times" rule, Iggy's .087 mils suddenly become .87 of designed-for error, which is a lot more than _I'd_ be willing to tolerate -- 2000 counts would give a designed-for error of 0.3 mil, which isn't lovely, but is probably getting limited more by the machine than the encoders (which is what you aim for). Those sinusoidal encoders will give you essentially infinite resolution, so you'll never have to wonder if you're losing accuracy to quantization noise. You can still lose accuracy in lots and lots of other ways, but not to quantization... -- Tim Wescott Control system and signal processing consulting www.wescottdesign.com |
#19
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Ignoramus21167 fired this volley in
: Your accuracy number for my example was 0.02". If you increase the count by 3 times (as in your mention of 2,000 CPR) then the accuracy only improves to 0.006", also an unacceptable number. Something is not right. Thanks for doing the math, Iggy. Bingo! You win the big teddy bear. If a servo misses re-positioning 128 (or 256!) times in a full stroke of the machine, it's undersized, or being driven too fast. I have a couple of machines that would upchuck if they detected TWO missed steps in 38". LLoyd |
#20
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
"Pete C." fired this volley in news:4c093eef$0
: I think the issue is encoder "lines" vs. "counts". The encoders on the machines I worked on were 2,000 line encoders, which as Karl noted could provide 8,000 counts per revolution. The US digital site uses counts, so presumably the 720 CPR encoder is 180 line? If, in fact, that's the case, then it's a huge difference. But still, that only gets you to roughly 11:1 better than the .02 error... and a thou in CNC work is NOT good enough! The thing about a servo missing up to 256 counts in a full-excursion run is nuts. My OLD R2E4 doesn't do that sort of thing. Newer, more capable servo drivers and faster encoders couldn't possibly miss that bad. LLoyd |
#21
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
"Pete C." fired this volley in news:4c094cf5$0$1092
: Lines = Number of physical lines on the encoder disk Counts = Lines * 4 (leading and trailing edges of the two quadrature channel signals) yep... and I have some cheap motors from old TI printers that have 720 LINE encoders. These are from the mid-80's, and they were not high-end expensive equipment like CNC mills or lathes. LLoyd |
#22
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
"Lloyd E. Sponenburgh" wrote: "Pete C." fired this volley in news:4c094cf5$0$1092 : Lines = Number of physical lines on the encoder disk Counts = Lines * 4 (leading and trailing edges of the two quadrature channel signals) yep... and I have some cheap motors from old TI printers that have 720 LINE encoders. These are from the mid-80's, and they were not high-end expensive equipment like CNC mills or lathes. LLoyd The resolution (within reason) isn't the expensive part, the rugedizing to survive a machine tool environment is. |
#23
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
"Lloyd E. Sponenburgh" wrote: "Pete C." fired this volley in news:4c093eef$0 : I think the issue is encoder "lines" vs. "counts". The encoders on the machines I worked on were 2,000 line encoders, which as Karl noted could provide 8,000 counts per revolution. The US digital site uses counts, so presumably the 720 CPR encoder is 180 line? If, in fact, that's the case, then it's a huge difference. But still, that only gets you to roughly 11:1 better than the .02 error... and a thou in CNC work is NOT good enough! The thing about a servo missing up to 256 counts in a full-excursion run is nuts. My OLD R2E4 doesn't do that sort of thing. Newer, more capable servo drivers and faster encoders couldn't possibly miss that bad. LLoyd +/- 128 counts isn't the same as missing by 256 counts. Some servo drives will have much tighter tolerance, and that is the limit under load, not normal. I used +/- 128 in the example since that is the max error of the popular Gecko servo drives, more than 128 off and it will generate an alarm to stop the control. |
#24
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
"Lloyd E. Sponenburgh" wrote: Ignoramus21167 fired this volley in : Your accuracy number for my example was 0.02". If you increase the count by 3 times (as in your mention of 2,000 CPR) then the accuracy only improves to 0.006", also an unacceptable number. Something is not right. Thanks for doing the math, Iggy. Bingo! You win the big teddy bear. If a servo misses re-positioning 128 (or 256!) times in a full stroke of the machine, it's undersized, or being driven too fast. I have a couple of machines that would upchuck if they detected TWO missed steps in 38". LLoyd You're confusing cumulative error in a stepper system with the non-cumulative following error in a servo system. A servo system needs fine encoder resolution so that the servo loop can function properly and hold position against loads. The max following error is the point that the servo will alarm and shutdown. If the system is designed correctly, 128 counts should translate into something like 1/10 of the nominal position resolution of the system, like 0.0001" if you expect to utilize 0.001" positioning. |
#25
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Karl Townsend wrote: I've had good luck with these: http://usdigital.com/products/encode...rotary/kit/e6/ I'd get a higher count per inch encoder. get one that gives you a very even count per inch movement number, like 20,000 or 50,000 or even 100,000. remember you get four counts per encoder pulse. get the differential option. use their wire and connectors. Don't be a cheap skate and use the PC 5 volt supply. get one just for the encoders. I usually get a multiple voltage out DC power supply. Seems like I always need 24, 10, 5 I don't know if EMC does analog (10V) or opto isolated digital I/O (24V) Karl Would you have any use for some of these? Integrated power designs 65 Watt switching power supply: Model # SRW-65-4006 4" X 6" X 1.25" +5V @ 5A / +24V @ 1A / +15V @ 2A / -15V @ 2A http://www.ipdpower.com/home/download.asp?id=379 The website is screwed up. Right click to save and add .pdf, then select all file types before saving. I still have 25 NOS units on hand for $25 each. -- Anyone wanting to run for any political office in the US should have to have a DD214, and a honorable discharge. |
#26
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Tim Wescott wrote: On 06/04/2010 10:04 AM, Pete C. wrote: Ignoramus21167 wrote: On 2010-06-04, Pete wrote: Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. Another one is to reuse a converter board from the existing Heidenhain controller. Same issue as above. (and I do have documentation, it just does not say what is what). I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: http://usdigital.com/products/encode...otary/kit/e7p/ They seem like they will fit, however, I am slightly surprised at the price. I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. i Those are pretty low resolution, only up to 720 CPR. The encoders on the CNC machines I used to work on were all around 2,000 CPR. You have to do the math with your ballscrew pitch to figure out what the final movement resolution is, but I expect it will be far too low. Also since this is a servo setup, the minimum servo error with such a low resolution encoder could equate to an unacceptable position error at the cutter. Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i .0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. The normal rule of thumb in servo design is that the encoder error should be no more than 1/10th the desired total system error. Your 128 count rule would mean that your 1/8" pitch ball screw would give an error of 128 * 0.125" / 2 / 2000 = .004". Is that really what you meant? Four mils error? The line / count confusion is there, so 2000 line = 8000 count, so more like .001" error to shutdown. The +/- 128 count is based on the alarm threshold in the popular Gecko servo drives, not the higher end stuff and also isn't the normal following error. 'course, by my "ten times" rule, Iggy's .087 mils suddenly become .87 of designed-for error, which is a lot more than _I'd_ be willing to tolerate -- 2000 counts would give a designed-for error of 0.3 mil, which isn't lovely, but is probably getting limited more by the machine than the encoders (which is what you aim for). Those sinusoidal encoders will give you essentially infinite resolution, so you'll never have to wonder if you're losing accuracy to quantization noise. You can still lose accuracy in lots and lots of other ways, but not to quantization... -- Tim Wescott Control system and signal processing consulting www.wescottdesign.com |
#27
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Would you have any use for some of these? Integrated power designs 65 Watt switching power supply: Model # SRW-65-4006 4" X 6" X 1.25" +5V @ 5A / +24V @ 1A / +15V @ 2A / -15V @ 2A http://www.ipdpower.com/home/download.asp?id=379 The website is screwed up. Right click to save and add .pdf, then select all file types before saving. I still have 25 NOS units on hand for $25 each. Great prcie but i need more amps @24V Karl |
#28
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Karl Townsend wrote: Would you have any use for some of these? Integrated power designs 65 Watt switching power supply: Model # SRW-65-4006 4" X 6" X 1.25" +5V @ 5A / +24V @ 1A / +15V @ 2A / -15V @ 2A http://www.ipdpower.com/home/download.asp?id=379 The website is screwed up. Right click to save and add .pdf, then select all file types before saving. I still have 25 NOS units on hand for $25 each. Great price but I need more amps @24V No problem. I just thought I owuld ask. They are low noise, and were being installed in some sensitive audio equipment. I may still have some good 8" floppy drive power supplies. They have more current at 24 volts. They usually have +5 & and adjustable +12 to +15 output that will go down to +10. -- Anyone wanting to run for any political office in the US should have to have a DD214, and a honorable discharge. |
#29
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
In article ,
Ignoramus21167 wrote: The servos on my mill have sinusoidal encoders which are not compatible with most CNC control stuff. I am looking at various options. One is to make a converter to convert sinusoidal to quadrature. It is a pain in the butt to do due to lack of documentation on what wire does what. What do you mean by "sinusoidal encoders"? Synchro transformers running at 60 Hz or 400 Hz single phase? What is the make and model of the encoders? If it's a synchro, there will be five wires (perhaps plus a safety ground). Two wires will go to a single coil on the rotor. Three wires will go to a Y or Delta sator coil. http://www.eugeneleeslover.com/USNAVY/CHAPTER-10-B.html http://en.wikipedia.org/wiki/Synchro Synchros are still used because they are simple and extremely rugged. Anyway, one can buy chips that will convert synchro to angle, and vice versa, and the wiring isn't that hard to figure out, so making a converter isn't that hard. Joe Gwinn |
#30
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
"Tim Wescott" wrote in message ... On 06/04/2010 10:04 AM, Pete C. wrote: snip .0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. The normal rule of thumb in servo design is that the encoder error should be no more than 1/10th the desired total system error. The 1/10th rule of thumb is what I always used, I hardley ever see an error of more than 2 counts at any kind of low speeds. I don't recommend trying ..0001 tolerance at rapid speeds. With 1 count error, my system slowly corrects, with 2 counts it corrects quite rapidly, you have to be on the ball to see it. I don't think I've ever seen 3 count error unless I use controls to see it. Your 128 count rule would mean that your 1/8" pitch ball screw would give an error of 128 * 0.125" / 2 / 2000 = .004". Is that really what you meant? Four mils error? If I got a 128 count error then I would tune or replace the control. 'course, by my "ten times" rule, Iggy's .087 mils suddenly become .87 of designed-for error, which is a lot more than _I'd_ be willing to tolerate -- 2000 counts would give a designed-for error of 0.3 mil, which isn't lovely, but is probably getting limited more by the machine than the encoders (which is what you aim for). Those sinusoidal encoders will give you essentially infinite resolution, so you'll never have to wonder if you're losing accuracy to quantization noise. You can still lose accuracy in lots and lots of other ways, but not to quantization... -- Tim Wescott Control system and signal processing consulting www.wescottdesign.com Sounds reasonable to me, I don't recommend feeding 250 ipm if you have .0001 tolerance unless you have a machine specifically designed to do this. I may get 10 count error in rapids but not on any speed I would cut at. I haven't even optimized the PID gains on my CNC lathe but it gets within 0.0001 within a second or so. RogerN |
#31
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Pete C. wrote:
Ignoramus21167 wrote: I have opened up one of my servos to see the shaft. It is a 10mm shaft. For that size, US Digital has suspiciously cheap encoders E7P: Well, on the back of the envlope calculation is: ball screw pitch 1/8" (0.125"). Pulley ratio 1:2. So, 720 CPR results in 0.125"/2/720 = .0000868 inch per cycle OR 0.002mm That would be acceptable to me. i .0000868" per count * servo error tolerance of +/- 128 counts (256 count window) = 0.0222208" error which is very significant and would not be acceptable to me. Where do you get this "servo error tolerance +/- 128?" That sounds like you are assuming the Gecko 320 series of drives, which Iggy will not be using. But, anyway, the US Digital encoders have caused all sorts of problems in a variety of applications, mostly due to a total lack of decoupling capacitors in the encoder. I have worked with the CUI AMT102 encoders, and they seem to work well. But, they only go up to 8 mm ID on the wheel adaptor. They have a DIP switch inside to set the resolution, and also have the index pulse. There is a cable adaptor that converts the output to differential. They are available from Digi-Key, and quite affordable, just over $30 with either the plain or differential cable. Jon |
#32
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Ignoramus21167 wrote:
I will check what Jon's PPMC does, but I think that he takes 5v differential signals. The PPMC has jumpers to set each encoder for differential or single-ended, and provides up to 200 mA at 5 V to each encoder with a self-resetting thermal "fuse" (PolySwitch). It also has a termination resistor (120 Ohms) when in differential mode to absorb reflections on the lines. Jon |
#33
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Joseph Gwinn wrote:
What do you mean by "sinusoidal encoders"? Synchro transformers running at 60 Hz or 400 Hz single phase? What is the make and model of the encoders? If it's a synchro, there will be five wires (perhaps plus a safety ground). Two wires will go to a single coil on the rotor. Three wires will go to a Y or Delta sator coil. These have incandescent LIGHT BULBS in them, so there is NO DOUBT, whatsoever, they are optical, and also OLD. LEDs came out in the late 1960s, and due to the safety aspects of encoder failures, this is one of the first applications they went into. So, these must be from the late 1960's or very early 1970s. Heidenhain made analog output encoders, that either produced a current or voltage output, but was not converted to digital signals in the encoder. The eclipsing of the light path by the disc caused the signals from the encoder to vary in an approximately sinusoidal pattern. It was possible to interpolate these signals to increase encoder resolution, but some systems used the basic resolution without interpolation. There are a LOT of these Heidenhain analog or sinusoidal encoders out there! Synchro transformers are fairly rare in shaft angle measurement in the machine tool business. A Scott-Tee transformer is needed to convert to quadrature-related signals. Simply winding the stator with two coils at 90 degrees provides the quadrature relationship directly without the Scott-Tee, and so is much more commonly used in this industry. That is called a resolver. They put a rotary transformer in there, too, to get rid of the brushes, thus making a "brushless resolver". These typically have 6 wires, but can be made with 5. Jon |
#34
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
Ignoramus21167 writes:
I Wanted to know if anyone has any suggestions, as I do not want to go a wrong way. I have been using HP (Agilent) HEDS 5645-H06 encoders for years with Geckodrives. Here is the datasheet: http://www.truetex.com/heds.pdf These are 400 CPR quadrature encoders, so you get a resolution of 1600 per revolution, which with a 2:1 pulley translates to 3200 per rev, which on a Bridgeport table with 0.2" pitch screws (ballscrews or leadscrews) gives you about 0.0004" resolution. Given that backlash and rigidity slop, even with ballscrews and belt drive, is much more than that, this is plenty of resolution and an excellent match to the Geckodrive methods. The 1/10 rule-of-thumb is arbitrarily costly, less reliable, and of little value, in my engineering opinion. There should be an "impedance match" between the digital control resolution and the mechanical tolerances, for optimal economy and performance. I have a surplus quantity of the new HP items for sale at $50/each. These are made for standard 1/4" encoder shaft mounts, but if you are equipped for metalworking you can make your own adapters for other size shafts and mounts. I've even inserted 1/4" dowel pins into non-servo DC motor shafts to convert them to inexpensive high-power servos: This page shows how I did this with the HP HEDS item: http://www.truetex.com/servomod.htm |
#35
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
On 2010-06-05, Jon Elson wrote:
Ignoramus21167 wrote: I will check what Jon's PPMC does, but I think that he takes 5v differential signals. The PPMC has jumpers to set each encoder for differential or single-ended, and provides up to 200 mA at 5 V to each encoder with a self-resetting thermal "fuse" (PolySwitch). It also has a termination resistor (120 Ohms) when in differential mode to absorb reflections on the lines. It looks like you have provided everything, and that with differential encoders, no additional noise suppression is needed. Neat. Thanks i |
#36
Posted to rec.crafts.metalworking
|
|||
|
|||
Encoders? Karl?
On 06/04/2010 06:48 PM, RogerN wrote:
"Tim wrote in message ... On 06/04/2010 10:04 AM, Pete C. wrote: snip snip again Sounds reasonable to me, I don't recommend feeding 250 ipm if you have .0001 tolerance unless you have a machine specifically designed to do this. Now _that_ would be a fun machine to do the control design on -- or a nightmare, depending on whether I came in at the "blank sheet of paper" stage or if I came in at the "we blundered through this without knowing what we were doing and it's all screwed up, how little can you do and have it all meet spec?" Granted, the latter case is often more lucrative, but no one comes out of it just thrilled with life. I may get 10 count error in rapids but not on any speed I would cut at. I haven't even optimized the PID gains on my CNC lathe but it gets within 0.0001 within a second or so. A second seems like a long time -- but then 0.0001 is a small error. You do realize that 0.0001 error in the measurement may be much bigger by the time you get out to the cutting tool? -- Tim Wescott Control system and signal processing consulting www.wescottdesign.com |
Reply |
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
Renco Encoders web site ? | Metalworking | |||
Ref:Hurco Encoders | Metalworking | |||
Encoders | Metalworking | |||
Rotary encoders for 65XX and 66XX power supplies | Electronics Repair | |||
Connecting encoders | Metalworking |