THK KR15 cheap on eBay

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

ray_parkhurst
Posts: 3417
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

lothman wrote:
ray_parkhurst wrote:Well, I went ahead and built a proof of concept of my drawing. I made a few changes:

1) used an 80mm long THK SRS9GM instead of the 100mm long 12mm rail. This forced me to make a 5mm shim instead of a 2mm one. I made a simple square-ish shim that fits between the mounting screws.
Why don't you use a THK rail with two ballbeared sledges like the KR2602B (the "B" is for the two sledge model) . OK this version has 2mm pitch but is much stiffer than the smaller versions. So you don't have to bother about alignment but on the other hand loose some of the stroke, since there is an additional sledge in the rail.
That is of course an option, and I've talked about it in other posts. The idler rail has the same basic functionality. My motivation for posting the KR15 was their availability for very low cost with motor mount, shaft joiner, and stepper included. I expected folks would use the KR15's as I do for moving subjects, but several folks have purchased them with expectation of moving camera/bellows/lens, and the idler gives them a fairly cheap and flexible way to make this work.

JohnyM
Posts: 463
Joined: Tue Dec 24, 2013 7:02 am

Post by JohnyM »

Well, my pair will (hopefully) turn into microscope stage :)
I find 30mm travel too small for anything else than micro (mine starts from 10x)

viktor j nilsson
Posts: 423
Joined: Fri Mar 01, 2013 1:43 am
Location: Lund, Sweden

Post by viktor j nilsson »

ray_parkhurst wrote:
lothman wrote:
ray_parkhurst wrote:Well, I went ahead and built a proof of concept of my drawing. I made a few changes:

1) used an 80mm long THK SRS9GM instead of the 100mm long 12mm rail. This forced me to make a 5mm shim instead of a 2mm one. I made a simple square-ish shim that fits between the mounting screws.
Why don't you use a THK rail with two ballbeared sledges like the KR2602B (the "B" is for the two sledge model) . OK this version has 2mm pitch but is much stiffer than the smaller versions. So you don't have to bother about alignment but on the other hand loose some of the stroke, since there is an additional sledge in the rail.
That is of course an option, and I've talked about it in other posts. The idler rail has the same basic functionality. My motivation for posting the KR15 was their availability for very low cost with motor mount, shaft joiner, and stepper included. I expected folks would use the KR15's as I do for moving subjects, but several folks have purchased them with expectation of moving camera/bellows/lens, and the idler gives them a fairly cheap and flexible way to make this work.

Yes, all the blame for using this rail in a sub-optimal fashion must fall on the cheapskate that didn't want to pay for a full-sized rail (=me).

I mostly want to shoot very small flies, 2-5mm body length, or parts therof, so 30mm is plenty long for me.

Viktor

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Update on KR15

Post by mawyatt »

Folks,

A few things you might want to know before using these superb US Automation KR15 based rails that Ray has shown us.

The USA internal controllers/electronics are very nicely done and involve a separate motor driver chip and main controller chip. A switch mode power converter is used for efficiency, other chips create the RS485 interface support electronics. I'm sure these will work well if one has a RS485 interface, which I don't. So I've elected to remove the electronics and drive the NEMA 11 stepper motor directly.

I've created a special test bed for these USA KR15 rails/motors with the new stack & stick controller I'm developing, so I've put them thru various tests to evaluate their performance. The USA specs are not very clear on the actual motor current ratings, the actual NEMA 11 motors windings (wire size) can handle large currents (I've disassembled the motor to investigate). However, the case can get very warm when running high currents from a high supply voltage (~12V). After some experimentation keeping the motor current below 0.5 amp in the motor driver control seems like a good idea, the KR15 rails/motors perform well and the case becomes slightly warm. The USA specs show 0.7 amps, but this is not just the motor current but the driver/control electronics, and the specs show a 6.2 watt dissipation from 24 volts (258ma), so it's uncertain what the actual motor specs are. The windings can handle 0.5 amps and the case doesn't get hot, so that's seems like a good current limit.

The rails have just over 30mm of travel and I created a test where the rail is driven within 1mm of each KR15 stop (nice rubber bumper) end. This requires 48000 steps (30mm* 1mm pitch * 200 motor step * 8 micro steps). The motion control algorithm is somewhat complex, but basically ramps up the motor rate to a maximum rate then ramps down to reach the desired position. These were at a relatively brisk rate and I even tried to mess up the rail & controller by issuing dynamic motion reverses & changed positions on the fly and not allowing the rail to achieve "rest" (basically jerking it all over the place !). I've even removed the motor drive, turned off the 12V, and even removed (unplugged & replugged) the motors, abuse squared!! After many hours of testing I can say these rails/motors performed superbly, and I couldn't determine a single missed step or position error in spite of my attempts to mess things up! Must admit I'm just looking at the final zero position of the stage and motor shaft position to determine if something got messed up, later when this is attached to a proper optical setup I'll be able evaluate much more detail. Overall I would say these USA KR15 with the controller worked beautifully, extremely smooth, quick, quite and precise.

A couple last cautions, many stepper controllers available do not allow user motor maximum current control, either by pot or software. These could potentially overheat the KR15 motors.

Also the KR15 have a very shallow recess on the base mounting bolt holes (so do the KR20s, but it's deeper at 3mm), it's only ~2mm. Most standard mounting bolts have heads that are higher than 2mm and will interfere with the carriage (not a good thing!). Make sure you use shallow head mounting bolts with 2mm heads or less.

I'm mounting one USA KR15 on top of another for X & Y positioning. If you remove the aluminum plate on the USA rail the 4 hole pattern has the long axis at 14mm, which conveniently matches the KR15 mounting hole base 14mm spacing. I ground down a couple M3 bolts so the heads were <2mm and mounted the 2 USA KR15s together, works beautifully!! Sure it's not the "best" mechanical solution (custom plate is better) but for my intended purpose to support a small, light subject this should work just fine!!

Anyway, hope this helps those interested in the USA KR15 rails.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

viktor j nilsson
Posts: 423
Joined: Fri Mar 01, 2013 1:43 am
Location: Lund, Sweden

Post by viktor j nilsson »

ray_parkhurst wrote:. Make sure when you buy the controller that you get a motor control cable like I have shown. The one shown came from Peter so he should be able to send one with the controller. The other end of the cable plugs into the mjkzz controller.
Anyone knows how to contact Peter (mjkzz) directly? I want to make sure I get the motor cable when I order the controller. I've tried to ask through the contact form on the homepage and I've tried to PM him on here, but I don't think that either is working.

Pau
Site Admin
Posts: 6053
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

Viktor, check your PM inbox
Pau

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

viktor j nilsson wrote:
ray_parkhurst wrote:. Make sure when you buy the controller that you get a motor control cable like I have shown. The one shown came from Peter so he should be able to send one with the controller. The other end of the cable plugs into the mjkzz controller.
Anyone knows how to contact Peter (mjkzz) directly? I want to make sure I get the motor cable when I order the controller. I've tried to ask through the contact form on the homepage and I've tried to PM him on here, but I don't think that either is working.
viktor,

I very briefly (experiment) ran the USA KR15 with NEMA 11 motor from 12V with ~1.5a, it became very hot quickly, then I dropped the current ~1a and still too hot for my liking, so I ended up using ~0.5a as mentioned. At this current the motor seems very happy and delivers nice performance. These small NEMA 11 motors do not have much surface area as the larger NEMA 17 motors to get rid of the internal heat generated .

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 3417
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

mawyatt wrote:I very briefly (experiment) ran the USA KR15 with NEMA 11 motor from 12V with ~1.5a, it became very hot quickly, then I dropped the current ~1a and still too hot for my liking, so I ended up using ~0.5a as mentioned. At this current the motor seems very happy and delivers nice performance. These small NEMA 11 motors do not have much surface area as the larger NEMA 17 motors to get rid of the internal heat generated
What are you running for idling current? For horizontal setups, almost no idling current is needed, and in reality for these 1mm pitch shafts, very little idle current is needed even for vertical setups with reasonable loads.

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

ray_parkhurst wrote:
mawyatt wrote:I very briefly (experiment) ran the USA KR15 with NEMA 11 motor from 12V with ~1.5a, it became very hot quickly, then I dropped the current ~1a and still too hot for my liking, so I ended up using ~0.5a as mentioned. At this current the motor seems very happy and delivers nice performance. These small NEMA 11 motors do not have much surface area as the larger NEMA 17 motors to get rid of the internal heat generated
What are you running for idling current? For horizontal setups, almost no idling current is needed, and in reality for these 1mm pitch shafts, very little idle current is needed even for vertical setups with reasonable loads.
This is in my experimental setup for evaluating various motor controllers, so differentiating between idle and active isn't a condition yet. Since this is a reactive energy restoring effect (motor inductive currents are recycled back to the supply), the total supply current is much less than the actual motor coil currents. I can vary the motor coil current limit with PWM by software control, so these are the software settings. Others may not have the control setup that I'm using and need to watch the motor currents to prevent excessive motor heat built-up.

When I get things working I'll be able to change the current limit, microstep size, acceleration profile , deceleration profile, and speed on the fly without losing a step. I'll have a motor/rail profile for each rail and motor combo to help optimize the setup, this profile will get loaded up at start. So the controller will be able to handle various setups with various combinations of things.

As of now I've got the USA KR15 with this new controller (only doing single axis for starters) working better than anything I've seen or used. It's so smooth, precise and quite, but surprising fast (covering 30mm in a few seconds). Must admit after initially using the TI DRV8825 I didn't think I would be able to achieve this level of performance.

Still have lots of work to do, my coding skills are getting a little better, so things aren't quite as painful to do, and hopefully I don't need to spend any more serious $, already spent way more than a new super quality lens :shock:. But learning new stuff isn't always cheap :roll:

Edit: Just measured the motor coil DCR and it's ~5.6 ohms, so with a rms/avg coil current of 1.5a this becomes 12.6 watts per coil or 25.2 watts total. That's a lot of power for the small NEMA 11 case to dissipate. At 1a it's 11.2 watts, still a lot, but at 0.5a it's only 2.8 watts.

I measured the 12V supply current at different controller current limit settings.

Current Limit Setting[]Measured Static[]Dynamic
990ma[]561ma[]602ma
496ma[]159ma[]186ma
343ma[]97ma []121ma
174ma[]49ma []68ma
1ma []23ma []33ma (actually works, no detected missed steps!!)

More on this current limit and efficient inductive recycling mode. I did run the motor very briefly (don't want to do any heat induced damage, so not going above 1a again) at 1452ma current limit setting. I got some strange results so didn't initially post, now I've had time to think about what happened I'll elaborate.

The 12 volt static current went to ~1700ma and dynamic current dropped to under 1500ma. The voltage drop across the motor is comprised of the I*R drop and the L*di/dt drop. The H bridge driver actually has two voltage drops, the current sinking drop (low side) and the current sourcing drop (high side). The sum of these drops and the motor drop can not exceed the supply voltage, or the motor can't sustain the desired current. Well at 1452ma coil current, this would predict a resistive motor drop (I*R) of ~8.1 volts, with the driver sink and source drops this leave no voltage room for the inductive L*di/dt current and the motor falls away from efficient reactive current recycling. A sufficiently high supply voltage is required to "force" the current thru the inductive windings and with the large I*R drop there's not enough supply voltage. Larger supply voltages are desirable because they can "force" the inductive motor current and accelerate the motor quicker. So this helps expand why the supply current rapidly increases when the coil current limit is set above ~1a for these motors. Note that a 1~2 ohm motor would not behave this way, since the I*R voltage drop is much lower.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

mawyatt wrote:
ray_parkhurst wrote:
mawyatt wrote:I very briefly (experiment) ran the USA KR15 with NEMA 11 motor from 12V with ~1.5a, it became very hot quickly, then I dropped the current ~1a and still too hot for my liking, so I ended up using ~0.5a as mentioned. At this current the motor seems very happy and delivers nice performance. These small NEMA 11 motors do not have much surface area as the larger NEMA 17 motors to get rid of the internal heat generated
What are you running for idling current? For horizontal setups, almost no idling current is needed, and in reality for these 1mm pitch shafts, very little idle current is needed even for vertical setups with reasonable loads.
This is in my experimental setup for evaluating various motor controllers, so differentiating between idle and active isn't a condition yet. Since this is a reactive energy restoring effect (motor inductive currents are recycled back to the supply), the total supply current is much less than the actual motor coil currents. I can vary the motor coil current limit with PWM by software control, so these are the software settings. Others may not have the control setup that I'm using and need to watch the motor currents to prevent excessive motor heat built-up.

When I get things working I'll be able to change the current limit, microstep size, acceleration profile , deceleration profile, and speed on the fly without losing a step. I'll have a motor/rail profile for each rail and motor combo to help optimize the setup, this profile will get loaded up at start. So the controller will be able to handle various setups with various combinations of things.

As of now I've got the USA KR15 with this new controller (only doing single axis for starters) working better than anything I've seen or used. It's so smooth, precise and quite, but surprising fast (covering 30mm in a few seconds). Must admit after initially using the TI DRV8825 I didn't think I would be able to achieve this level of performance.

Still have lots of work to do, my coding skills are getting a little better, so things aren't quite as painful to do, and hopefully I don't need to spend any more serious $, already spent way more than a new super quality lens :shock:. But learning new stuff isn't always cheap :roll:

Edit: Just measured the motor coil DCR and it's ~5.6 ohms, so with a rms/avg coil current of 1.5a this becomes 12.6 watts per coil or 25.2 watts total. That's a lot of power for the small NEMA 11 case to dissipate. At 1a it's 11.2 watts, still a lot, but at 0.5a it's only 2.8 watts.

I measured the 12V supply current at different controller current limit settings.

Current Limit Setting[]Measured Static[]Dynamic
990ma[]561ma[]602ma
496ma[]159ma[]186ma
343ma[]97ma []121ma
174ma[]49ma []68ma
1ma []23ma []33ma (actually works, no detected missed steps!!)

More on this current limit and efficient inductive recycling mode. I did run the motor very briefly (don't want to do any heat induced damage, so not going above 1a again) at 1452ma current limit setting. I got some strange results so didn't initially post, now I've had time to think about what happened I'll elaborate.

The 12 volt static current went to ~1700ma and dynamic current dropped to under 1500ma. The voltage drop across the motor is comprised of the I*R drop and the L*di/dt drop. The H bridge driver actually has two voltage drops, the current sinking drop (low side) and the current sourcing drop (high side). The sum of these drops and the motor drop can not exceed the supply voltage, or the motor can't sustain the desired current. Well at 1452ma coil current, this would predict a resistive motor drop (I*R) of ~8.1 volts, with the driver sink and source drops this leave no voltage room for the inductive L*di/dt current and the motor falls away from efficient reactive current recycling. A sufficiently high supply voltage is required to "force" the current thru the inductive windings and with the large I*R drop there's not enough supply voltage. Larger supply voltages are desirable because they can "force" the inductive motor current and accelerate the motor quicker. So this helps expand why the supply current rapidly increases when the coil current limit is set above ~1a for these motors. Note that a 1~2 ohm motor would not behave this way, since the I*R voltage drop is much lower.

Best,
Update: Think I've found the motor (or similar) used in the USA KR15, it has a rated 670ma current with 5.6 ohms resistance and thus a 3.8 volt motor. The inductance is 4.2mH which makes the motor time constant 23.5ms, which is important in understanding how the PWM current control behaves. So the assumption of not operating at 1.5a, or 1a, and using 0.5a is probably a good idea!!

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 3417
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

mawyatt wrote: Update: Think I've found the motor (or similar) used in the USA KR15, it has a rated 670ma current with 5.6 ohms resistance and thus a 3.8 volt motor. The inductance is 4.2mH which makes the motor time constant 23.5ms, which is important in understanding how the PWM current control behaves. So the assumption of not operating at 1.5a, or 1a, and using 0.5a is probably a good idea!!
I run the KR15 NEMA-11 at 625mA running, 125mA idling. I have some MachMotion NEMA-11 that I run at the same conditions and they work great, no noticeable heating.

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

ray_parkhurst wrote:
mawyatt wrote: Update: Think I've found the motor (or similar) used in the USA KR15, it has a rated 670ma current with 5.6 ohms resistance and thus a 3.8 volt motor. The inductance is 4.2mH which makes the motor time constant 23.5ms, which is important in understanding how the PWM current control behaves. So the assumption of not operating at 1.5a, or 1a, and using 0.5a is probably a good idea!!
I run the KR15 NEMA-11 at 625mA running, 125mA idling. I have some MachMotion NEMA-11 that I run at the same conditions and they work great, no noticeable heating.
With the current recycling using a 496ma coil current only creates a 186ma actual supply current, so the power is just 2.2W from a 12 volt supply. The static supply current is only 159ma (1.9W), but the motor holding current is actually ~496ma!!

If your 625ma is the supply current then with 12V that's still only 7.5W, but it's dynamic which is OK.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 3417
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

mawyatt wrote:
ray_parkhurst wrote:
mawyatt wrote: Update: Think I've found the motor (or similar) used in the USA KR15, it has a rated 670ma current with 5.6 ohms resistance and thus a 3.8 volt motor. The inductance is 4.2mH which makes the motor time constant 23.5ms, which is important in understanding how the PWM current control behaves. So the assumption of not operating at 1.5a, or 1a, and using 0.5a is probably a good idea!!
I run the KR15 NEMA-11 at 625mA running, 125mA idling. I have some MachMotion NEMA-11 that I run at the same conditions and they work great, no noticeable heating.
With the current recycling using a 496ma coil current only creates a 186ma actual supply current, so the power is just 2.2W from a 12 volt supply. The static supply current is only 159ma (1.9W), but the motor holding current is actually ~496ma!!

If your 625ma is the supply current then with 12V that's still only 7.5W, but it's dynamic which is OK.

Best,
495mA hold current is probably much higher than is needed, though the resulting 1.4W of continuous dissipation should not cause a huge problem on this motor.

ray_parkhurst
Posts: 3417
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

I received my 200mm Arca "plate" and installed it onto the 3/8" carriage. I also installed the modified bellows I discussed earlier (or maybe in another post), and the overall structure is extremely rigid and smooth operating. Here's what it looks like:

Image[/img]

viktor j nilsson
Posts: 423
Joined: Fri Mar 01, 2013 1:43 am
Location: Lund, Sweden

Post by viktor j nilsson »

Looking very good, Ray. I must have missed the thread about your bellows, must check that out. I have one of those huge old Pentacon M42 bellows. In general I'm ok with it, but the front leg protrudes so far forwards that I sometimes bump into my xyz specimen holder. I'll probably change it for something else at some point.

My project is slowly moving forward. Got the controller and complementary motor cable from Mjkzz yesterday (thanks Peter!). The KR15 arrived a week ago. It is indeed very small, but with a very nice, solid feel to it. Still waiting for the linear guide rail from China. But in the meantime I'll try to get the KR15 operational.

Post Reply Previous topicNext topic