Linear actuator using ClearPath Integrated Servo System @ 72V DC

They are 56C frame motors.

Here is a link to purchase them directly from Teknic @ $799 each

https://www.teknic.com/model-info/CPM-MCPV-N0561P-RLN/

The gearbox I'm using is an 80:1.but may go to a 60:1 depending on the speeds achieved.

900USD is not bad, but then yes, you need to add the cost of the gearboxes, and other metal pieces for the 6DOF platform... The Clearpath motors are really great since its very easy to configure them to do exactly what you need via PC software utility.


For comparison I finished the development of the AMC-AASD15A servo controller that uses the inexpensive AASD-15A drives and servomotors. I calculated the cost for complete set servomotor and linear actuator to be from 700USD to 900USD. But these servo drives, being so cheap, they make a whining noise while on operation (10khz), which is not ideal to hear all the time...









Thanks
Thanos

@Nisch
Thanks for the info regarding the motors. Do you have such info for the gearbox as well ? It is a worm gear right ?

The gearboxes are wormdrive, 80:1, 75mm center distance, 1.25" shaft output, 5/8" shaft input.

Hey there! I've been following along some on this project and I've got an itch to build an actuator like this(when money permits). I made an account because I intended to ask what power supply you plan on using? So far I have only found the one provided by teknic that outputs 75vdc and the ones I can find for 72vdc are mostly chargers without a lot of amp output.
Thanks! Looking forward to seeing your project to completion.

Hey there

Sorry, haven't checked in here for quite a while. I was busy writing my software for the past ~10 months. Now that that is almost done, I'm back with my actuator build

Concerning your question: I am still not sure wether to use 6x Teknics IPC5 power supply (expensive!) , find another external power supply (a little less expensive), or simply use 6x12V car batteries to create 72V DC. If someone has some thoughts on those ideas, let me know

Dirty

Thank you @Nisch for sharing your experience with those motors. Your posts were the straw that broke the camels back :-D

I have finally decided to ditch my MCVC for MCPVs I have designed pots, ran belts up and down my actuator, wrote a firmware for the controller,... and in the end it just became so TOTALLY obvious that I should just switch to MCPVs to get rid of all of that. I wish I had better understood the difference between the MCVC and MCPV when I bought them a year ago, but I guess there is no use crying over spilled milk.

Anyways, I ordered 1 MCPV 3432 D-RLN for testing and if all goes as expected I will order the other 5 in a few weeks.



A few more changes I made on my actuator:

I might replace the BK12 bearing blocks with FK12 flange type bearing blocks. They are just as inexpensive and distribute the pressure more evenly onto the lower aluminum plate.


Another option I am testing out is going from an "all-in-line direct drive" configuration to a belt drive with the motor mounted on top.

It has a couple of advantages:

1. It would shorten the overall length by almost 200mm (8").
2. The additional space can be invested into longer ball screws. I am now using 700mm ball screws instead of the 500mm I used before. That increases the usable travel from 360mm to 560mm. All the while the actuator keeps the same overall length.
3. I am testing SFU 1610 ballscrews. Those have 10mm lead instead of 5mm. So I can get either twice the speed (~660mm/s ) or I can use a gear ratio of 2:1 and achieve the same speed at half the rpm (--> less noise). Or any solution in between by changing the belt drive.

Something else I experimented with is the optimization of the motion workspace:

...so far I have 1142mm lateral and 653mm vertical travel available and it allows me to play around with different configurations. All I need to do is plug in the 6 defining parameters:


The model gives me this "hull body" that represents all possible translatoric positions. If someone is interested in the formulae or simply wants to try it out, PM me, I can send you the Fusion 360 file.

Cheers,... Dirty

The “translational umbrella” is fantastic. Maybe something @pmvcda could implement in Mover so you could see what’s the best compromise for heave offset

Good stuff Dirty! I'm going to be needing to figure out the best dimensions and actuator placement for my new sim as well so I'll be looking for some help figuring it all out. I'm hoping to spread out the attachment points a bit from a rig like FlyPT's to better accommodate having full cockpits on the sim.

I was thinking the same thing about the motor placement based on actuator designs I've seen, I'm fairly certain I'll be installing them that way as well.

It looks like what you are offering up there is what I need to help me figure out the best dimensions for my sim. I've started learning some extremely basic F360, so send it on over Dirty and I'll see if I can figure it out. =D

I'll be working with the following design goals:

• A 6DOF Stewart with a linear actuators having a minimum of ~450mm travel (I'd love to get more)
• Sim must fit in a 7 foot (2.13 meter) by 11 foot (3.35 meter) space (the actual space is a few inches bigger than that)
• The largest practical upper frame (cockpit) area without making big sacrifices, though I don't mind sacrificing the range of angles the sim can produce.

I'm interested in smoothness and really big heave travel above all else. My servos are 1.3HP (1KW) each so no shortage of power. With 10mm pitch ballscrews my actuators should give me around 425 mm/sec.

So far, I only have a mathematical model for linear actuators. Crank arm setups are quite a bit more complicated At least I have no analytical solution there.

Hey Trip

Here's the link to the Fusion360 project. See if you can download it. Then you click on "Modify" --> "Change Parameters"...

...and enter only the 6 "User Parameters" under "Favorites":


When you hit OK, the model is being created parametrically.

You can use the measure-tool (hit "I") to measure the body from corner to corner, or the "section analysis" to get cross sections.


The colored dots represent the lower connecting points of your actuators, but they are not depicted in their ACTUAL positions. These positions are the positions where the lower connecting points WOULD end up, if you shifted the actuators parallel so that their upper connecting points ended up exactly in the center of the upper platform. Yah,... not the best explanation,... but correct none the less

The colors of the umbrella tell you which actuator is the limiting one when the center of the platform approaches that particular surface.

Let me know how it goes.

Dirty

Can anyone tell me, if this is a sensible setup to create a reliable power supply.



Would you recommend diodes? If so, where and which?

My thoughts behind this were: Mainly the batteries will power the rig with 72V DC (maybe a little more). As soon as the batteries' voltage drops below 75V the power supply will feed additional current and recharge the batteries.

Will this work?

Assuming lead acid batteries,
I think (but not sure) that 12V batteries need a min 12.9v to drive the lead acid chemistry in the charge direction.
so 6 x 12.9=77.4 V to begin minimum charging (Is that what they call trickle charging?).

A charged lead acid battery will rest around 12.5V x 6 = 75V
So your batteries will initially supply power and discharge until they approach the power supply of 72V and sit there for the short term.
If the power supply cant keep up in high demand and the voltage sinks, the batterries will deliver but this swill keep eating into the battery and the battery will not recover.
Over time, as the batteries self discharge and degrade, the power supply will deliver all the power even during the high demand spikes.
But I guess the batteries are there to abosrb regen current and perhaps they will still perform this function at the lower capacity state. But Im not so sure of this.

Perhaps the power supply could be set to 78V if your other system componenets can cope with it.
This may keep the batteries topped up.

Im not 100% on the above comments, happy to hear other thoughts.
I went through the same exercise in my 24v system with Sabertooth.
I've set the pwer supply to 26V to hopefully extend the life of the batteries.

All of this is the reason I decided batteries and power supply are a PITA and I'll avoid a system that requires batteries on my next do over.

Thanks Dirty, I will check it out as soon as I find a little time.

Hey @JAD, Thank you very much for the great explanation!

Makes total sense. And yes, it is quite a PITA that I simply never had on my radar. I thought 'How hard can it be to get 75V DC?'. ClearPath does provide 500W (1KW peak) DC power supplies, but with tax and shipment each will be around 250€. So ~1500€ just for the power supplies :-/ On top of ~3600€ for the motors themselves.

Compared to these guys...
https://de.aliexpress.com/item/32843516422.html?spm=a2g0s.9042311.0.0.7eb44c4dO0j57W

...where you'd get 6 for 1600€ (including tax & shipment), that plug directly into ordinary german 240V household wall power outlets, that is quite a difference. I have ordered Thanos AASD-15A board and will look into those AASD servo motors.

Didn't think I was ever gonna say this, but I might actually change horses in the middle of the stream.

Dirty

It’s hard to believe these Chinese specs.

My Clearpaths are rated with 2.8Nm for 1000 RPM continus torque vs. AASD claims to have 4Nm with 3000 RPM.

Are they 5 times stronger ??

They seems to have similar size. Well well ... I would like to know how the torque curve decreases with RPMs.

I would like to see a graph on torque vs. RPM as well. The folks on Thanos' Discord say that their AASD-15 actuators seem to be super powerful but that's not relative to anything and basically subjective at that.

I found this, but I have questions from my brief look at it. Can you guys take a look?

I notice it looks like they list my 4025 motors in teh 130st category, but I'm using 80st which is suspicious. I have to get out the door in a moment so I won't get a good look at this until much later tonight or tomorrow.

http://www.knd.com.cn/knd/upload/accessory/20122/2012220141540788030.pdf

Yes, this is correct. It provides full rated torque up to the max rated RPM...

Teknic for comparison




80st here :