Question New Build using Kollmorgan 12-943434C actuator

I found these actuators on Ebay and thought what the hell, for $265 for 6 or $44 per actuator after tax and shipping, this is a pretty good deal even if it's only to experiment and protype. I purchased 6 of them and proceeded to try to find out what I had just purchased. I bit of a cart and horse thing.

The seller didn't know anything, and the part number is a custom part number generated for the USPS and comes from a MDC sorting facility that was closed. No info on controls or what specs are on the actuator.

Kollmorgan responded to my email and evidently it is a IDC NH series with a 160v 2a PMDC motor with a brake on the end. The control drive can be generic as long as it meets the voltage/amps requirement. The official servo drives are the H3301B, H3321B, H3501.

Fortunately I was able to find some on ebay and was able to purchase all 6 drives for around $375.....Crap.
I figured I kinda screwed myself if I didn't get the drives since 160vdc power supplies seem to run $75 without having any controls built in. and these drives are super expensive.

My plan is to disassemble one of the actuators and measure the lead screw and should be able to deduce the full specs with the information I have. Then I should know the thrust, speed, stroke of the actuators, and confirm they pair up with the drives and function properly. The actuators appear to support reed switches/ hall effect magnets that get wired into the drive and will activate end-stop braking.

I'm wanting to build a Stewart Platform, that is as compact as possible, and setup more for space/flight sims vs racing.
My questions:

1. What is the best way to add position sensing? Is it necessary, or is homing and setting length and speed parameters enough?
2. What are my options for control interface? Thanos? SFX-100 shield? Any other viable options for DIYers?

Here are some pics

I just started bench testing these drives with IDC actuators. The problem I ran into is the acceleration/deceleration is great for a set speed point but when the speed input is changed it's like they use the slowest acceleration/deceleration setting. I even tried jogging, changing speed then jogging again and on the second jog it resumes at the first speed then decelerates over a few seconds to the second speed when the second jog was initiated. You can bench test with a couple of momentary switches and potentiometer.

I hope there is way to get the speed input response to honor the acceleration/deceleration setting pots but I don't know how to get support on 20 year old drives.

Use the wiring in Example 1: Jogging the Actuator from the drive manual

I assume this is some kind of safety feature and I could not find any configuration that respond quickly to remote speed input changes. I have begun looking at other drives for this reason.

ID/Kollmorgen actuators use hall or reed sensors mounted to cylinder.

Actually, there are a couple of different types of cylinder housings with different slot spacings, i don't remember the specific differences but can look later.

RPS-1 mechanical reed normally open
RPS-2 mechanical reed normally closed
RP1 hall effect normally open
RP2 hall effect normally closed

Also, 2 of my cylinders are belt drive and one is geared. The belt drive cylinders use HTD 5mm belts and pulleys, you could buy or 3d print some gears to change the ratio if you need to. I found a good timing belt length calculator online to determine what belt change I would need for a given pulley change.

I am testing with a arduino nano since I had them and they are code compatible with Uno using the updated SMC3 code. I am using an off the shelf pwm to 0-10v signal converter for the remote speed input, UNL2803 Darlington Array to boost logic signals to 12V and LM393B to buck 3301 drive output signals back down to 5v. The limit switches are connected to the drive and the drive output signals are connected to the arduino.

Nice, you have a good headstart on where I am at. I can't wait to see what you learn. I will update when the drives arrive and if I come up with anything new.

So the gear ratio is 3 to 1 on the one I opened, and the built in brake is 24vdc 10w. Which according to the manual I got, is not matching up. The manufacturer only lists 1to1, 1to1.5, and 2to1 gear ratios for their belt drive. The brake voltage is as 115vac in the book and on the drive.

I'm too green to post links.
Google "IDC N Series Actuator" and look at the kollmogern link pdf file
Google "IDC N2 Series Actuator" and look at the kollmogern link pdf file
Google "IDC EC Series Actuator" and look at the kollmogern link pdf file

Note that the definition of 2B and 5B swapped at some point (0.5", 0.2" lead)

The motors are rated 160V 2A? What's the gear ratio and pitch of the screw? To test what speed you can expect, connect a 12V battery or bench supply to the motor and time how long it takes to move from one end stop to the other. At 120V you can expect 10 times the speed and at 160V ~13 times...
UHU servo drives are very simple and accept step/dir input.

The specifications are all imperial based so nothing lands in mm properly.
Stroke: 10.25"
1 revolution distance: .1875"
Gear Ratio: 3:1
Motor RPM: 3900
Speed= [(3900/3)/60]*.1875=4.0625"/sec
Which works out to 103mm/sec.

So the theoretical speed is below 200mm/sec. From what I've read, anything below 200mm is kinda slow, is that correct? So I probably need to change gearing if I want fast enough response.
2:1 = 155mm/sec
1.5:1 = 206mm/sec

So at this point I will test the speed in real word results and look for a 12t timing gear

100mm/s is indeed a bit slow. 200mm/s would be OK if you had a 2DOF or 3DOF seat mover where the actuators are relatively close to the center point. But for a 6DOF rig the actuators are usually arranged in a ring around the rig with >1m diameter. Most people use actuators with around 600mm stroke and 300 to 500mm/s speed.

You could probably get away with a modified gear ratio of 1.5:1 if you place the actuatorsunder the rig and as close to each other as possible. The main problem is the small pitch of the screws. You can't run them much faster that ~2000RPM. 200mm/s requires 2500RPM which might work but it will make quite some noise. Unfortunatelly, it's not easy to change the pitch.

This helps me understand the design choices I see on other rigs. Such as using gearboxes and levers to achieve the speed and range of motion desired.

I've been rethinking my design, my current rig is a seat mover, with yaw. I like Race@Home's Low Profile surge, traction loss platform.

Woh-wah! Harrumph! So am feeling kinda dumb and overwhelmed. I am not an electrical engineer that's for sure.

The actuators are fine and I can get the gear ratio adjusted to within the mm/sec range.

The problem is the servos and drives are really dumb. The drives allow for 2 speed options and not feeding a pwm signal to the driver for speed, direction, etc responding to telemetry from a motion sim.

I can't figure out how to configure a PLC and interface that plays well with sim tools. After reading the issues that Thanos and Race@home have had developing an interface controller, I realize I am still not an electrical engineer! So probably need to sell the actuators paired with a driver, a salvage my loss. And look at using servos compatible with interfaces that have had all the impossible work done by peeps like Thanos.

I'm still reworking my rig to move in the axis I want, and to fix problems I have with it. But the servos will likely need to be compatible with established controllers.

Man you called it, these drives are not easy to adapt for this purpose. I explored several options ideas, and basically the only this I came up with is to use the drives only as 160vdc power supplies and use a different drive as a motion speed control. The problem is that any controller needs to be able to talk to Simtools and this elevates the expense to the point where it's more cost effective to use tried and true servos and drives that all that work has been done already.

The sticking point is that damn voltage, you can find a ton of generic drivers that you could use Arduino to control, but 160vdc@2a is not common, and those drivers are expensive and have proprietary software.

Not sure if this has been your experience or not.

I think I might keep 2 or 3 actuator/drive pairs, and try and sell the rest to someone wanting to build a CNC project. Or some other motion project.