Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Possible to use an industrial VFD to control an AC synchronous drive motor?

  • Please log in to reply
18 replies to this topic

#1 TallTanBarbie

TallTanBarbie

    Explorer 1

  • -----
  • topic starter
  • Posts: 85
  • Joined: 20 Mar 2023

Posted 15 June 2023 - 10:18 PM

Hello -- I have an old telescope mount with AC synch motors, and there isn't physical space and/or clearance to take out the AC drive motors and replace them with stepper motors, so I have to go with what I've got.

 

Like a lot of classic mounts, this one has 115v permanent magnet motors that run at either 1 or 1/2 RPM.

 

Have any of you used an industrial variable frequency drive (VFD) motor control unit to control one of these motors to a computer using MODBUS commands via RS485 ?

 

I don't need the telescope to slew, just fine motion control with a joystick N-S / E-W up to 4X or 8X of the sidereal rate.

 

I can do the programming, all I need is a way to control the AC frequency power to the motors, and if you can get a device like a VFD that can be controlled from a computer to do it, that would make this much easier, thanks :)

 

 



#2 Matthew Paul

Matthew Paul

    Viking 1

  • -----
  • Posts: 501
  • Joined: 06 Jun 2016

Posted 16 June 2023 - 06:02 AM

I have limited experience with setting up systems like this but when I worked in manufacturing the standard 120v single phase input converters were used to run three phase ac motors. I did not think that the controllers could vary the speed of a single phase ac motor. These were high load several HP applications. 


Edited by Matthew Paul, 16 June 2023 - 06:03 AM.


#3 Ed Jones

Ed Jones

    Aurora

  • *****
  • Posts: 4,617
  • Joined: 06 Apr 2004
  • Loc: Sin-sin-atti

Posted 16 June 2023 - 06:33 AM

Did you not see this thread?


  • TallTanBarbie likes this

#4 TallTanBarbie

TallTanBarbie

    Explorer 1

  • -----
  • topic starter
  • Posts: 85
  • Joined: 20 Mar 2023

Posted 16 June 2023 - 12:33 PM

Did you not see this thread?

No, I did not -- until just now, thanks very much.



#5 TallTanBarbie

TallTanBarbie

    Explorer 1

  • -----
  • topic starter
  • Posts: 85
  • Joined: 20 Mar 2023

Posted 17 June 2023 - 11:16 AM

After lots of searching, it seems there's only one single phase VFD currently on the market that can output 115v AC:

 

https://www.grainger...ive-115V-20RA39

 

BrandEATON

Manufacturer Part NumberDC1-S17D0NN-A20CE1

Enclosure RatingNEMA 0

Maximum Output Power1/2 hp

Input Voltage115V

ACInput Phase AC

SingleInput Frequency50/60 Hz

Output Phase AC

SingleOutput Voltage115V AC

Output Frequency Range0 to 500 Hz

Maximum Output Current7 A

With Bypass / No Bypass

No BypassWith Line Contactor / No Line Contactor

No Line ContactorStart/Stop ControlProgrammable

 

 

There's a lot of Arduino projects out there that people have made for this, but I'm not looking for control via directly wired analog buttons or switches, but instead from a computer sending commands over a network.

 

 

Whatever tells the motors to speed up or slow down has to be able to get a message from a remote machine.


  • Matthew Paul likes this

#6 TallTanBarbie

TallTanBarbie

    Explorer 1

  • -----
  • topic starter
  • Posts: 85
  • Joined: 20 Mar 2023

Posted 13 September 2023 - 09:51 AM

After lots of searching, it seems there's only one single phase VFD currently on the market that can output 115v AC:

 

https://www.grainger...ive-115V-20RA39

 

BrandEATON

Manufacturer Part NumberDC1-S17D0NN-A20CE1

Enclosure RatingNEMA 0

Maximum Output Power1/2 hp

Input Voltage115V

ACInput Phase AC

SingleInput Frequency50/60 Hz

Output Phase AC

SingleOutput Voltage115V AC

Output Frequency Range0 to 500 Hz

Maximum Output Current7 A

With Bypass / No Bypass

No BypassWith Line Contactor / No Line Contactor

No Line ContactorStart/Stop ControlProgrammable

 

 

There's a lot of Arduino projects out there that people have made for this, but I'm not looking for control via directly wired analog buttons or switches, but instead from a computer sending commands over a network.

 

 

Whatever tells the motors to speed up or slow down has to be able to get a message from a remote machine.

UPDATE:

 

I was able to get one of these VFDs and an AC Hurst 3002-001 synchro motor that runs at 1 RPM (similar to what many old telescope drives have) — and I got it to work smile.gif

 

This is for upgrading an existing 16" telescope mount that uses AC synchronous motors on both the RA and DEC axes.  The old drive correctors for these are impossible to find (especially a dual-axis AC unit).

 

Now I have to figure out how to get a computer to talk to it via RS-485 Modbus or use Arduino to emulate the external switches and potentiometers to adjust the frequency.  If I can't get this to work, plan B is to put in stepper motors with new gearing.

 

 

 

Hurst Synchro Motor Test Videos

 

 

https://mega.nz/file...csXHWM_cbz6eDYk

 

https://mega.nz/file...-zvyg3eDEjcl_CI


Edited by TallTanBarbie, 13 September 2023 - 09:53 AM.


#7 brave_ulysses

brave_ulysses

    Surveyor 1

  • -----
  • Posts: 1,505
  • Joined: 19 Apr 2009
  • Loc: far outside the wire

Posted 14 September 2023 - 08:33 PM

it appears on the huanyang 220v input vfds you can set PD008 to 110 for 110v output voltage. haven't tried it...



#8 luxo II

luxo II

    Voyager 1

  • *****
  • Posts: 11,560
  • Joined: 13 Jan 2017
  • Loc: Sydney, Australia

Posted 15 September 2023 - 04:50 AM

If the motors on your scope are small synchronous ones like those in old orange C8's and quite a few other telescope drives, you may be disappointed - their speed range is quite limited.

 

Many years ago as a young electronic engineer I designed and built a few drive correctors. At one point I tested the sychronous motors from a C8 on a lab bench, typically they could handle 0.5X the nominal rate, but with the usual load they wouldn't go much past 2X the nominal rate without stalling - only sufficient for fine slewing and guiding - not enough for fast slewing. Unloaded they will go quite a lot faster, but the trouble with these is the stall torque drops very quickly with increasing frequency.

 

I quickly ditched these for low-voltage steppers. A modern stepping motor can manage a far greater speed range - for example I used to choose gearboxes so that the nominal RA track rate would be around 10-15 steps/sec, and from that they could do more than 1,000 steps/sec, and if you use acceleration and deceleration, a 8,000 steps/sec is feasible with the right motor.

 

The one thing to note with steppers is that they draw maximum current (and have maximum torque) when stationary, so give some thought to reducing the current drawn when its not needed, if this is goingbto be an issue for a battery supply.


Edited by luxo II, 15 September 2023 - 04:54 AM.

  • don clement and Jaylab like this

#9 TallTanBarbie

TallTanBarbie

    Explorer 1

  • -----
  • topic starter
  • Posts: 85
  • Joined: 20 Mar 2023

Posted 15 September 2023 - 08:50 PM

If the motors on your scope are small synchronous ones like those in old orange C8's and quite a few other telescope drives, you may be disappointed - their speed range is quite limited.

 

Many years ago as a young electronic engineer I designed and built a few drive correctors. At one point I tested the sychronous motors from a C8 on a lab bench, typically they could handle 0.5X the nominal rate, but with the usual load they wouldn't go much past 2X the nominal rate without stalling - only sufficient for fine slewing and guiding - not enough for fast slewing. Unloaded they will go quite a lot faster, but the trouble with these is the stall torque drops very quickly with increasing frequency.

 

I quickly ditched these for low-voltage steppers. A modern stepping motor can manage a far greater speed range - for example I used to choose gearboxes so that the nominal RA track rate would be around 10-15 steps/sec, and from that they could do more than 1,000 steps/sec, and if you use acceleration and deceleration, a 8,000 steps/sec is feasible with the right motor.

 

The one thing to note with steppers is that they draw maximum current (and have maximum torque) when stationary, so give some thought to reducing the current drawn when its not needed, if this is goingbto be an issue for a battery supply.

I only wanted to do fine slew control with a joystick (NSEW).  I would like to use steppers, but there's so little room inside the housing for the RA motor I don't know if I could get a stepper motor to fit -- especially with a gearbox added.  And you're right -- the fastest speed the motor seemed to be happy with was 120hz, the slowest was 35hz.


Edited by TallTanBarbie, 15 September 2023 - 08:52 PM.


#10 luxo II

luxo II

    Voyager 1

  • *****
  • Posts: 11,560
  • Joined: 13 Jan 2017
  • Loc: Sydney, Australia

Posted 16 September 2023 - 06:07 PM

Well, a photo or two showing both axes - with a rule for scale - would speak a thousand words.

For steppers have you looked at www.pololu.com ?



#11 TallTanBarbie

TallTanBarbie

    Explorer 1

  • -----
  • topic starter
  • Posts: 85
  • Joined: 20 Mar 2023

Posted 16 September 2023 - 08:45 PM

Well, a photo or two showing both axes - with a rule for scale - would speak a thousand words.

For steppers have you looked at www.pololu.com ?

Hi, I've not heard of pololu.com, I'll check it out.

 

The mount is a Byers 812.

 

I don't want to alter it too much because it's a rare classic -- it has AC syncromotors on both axes -- but the AC RA axis motor is inside the main housing, and it's on a trapeze that can flip down to disengage the worm gear.

 

Got a message from another person on here about old Hurst stepper motors that have the same housing, found some for sale ( https://www.electron...er-12vdc-3-watt ), but the part number doesn't reference anything, so I can't find out the specs.

 

I wanted to see first if I could make some kind of control for it that would change the motor frequency based on joystick deflection, just fine slewing so I could adjust the position without having to hand move it.

 

Here are some pictures smile.gif

 

1693957275056

20230820 142107

1685987341420

1685987341362

20230605 124250

byers 812 Ra motor

byers 812 Dec motor

 

--Diana


Edited by TallTanBarbie, 16 September 2023 - 08:48 PM.

  • PrestonE likes this

#12 gregj888

gregj888

    Gemini

  • -----
  • Posts: 3,141
  • Joined: 26 Mar 2006
  • Loc: Oregon

Posted 16 September 2023 - 09:40 PM

Talltan,

 

Jomping threads, the AC controller can be made to run the syncros.  Just make sure the voltages match.  Tracking and guiding only.  Auto guide is trickier, probably easiest using the SST4 metodology and hard wire.

 

The pololu stepper stamps are what Onstep uses. Do yourself a huge favor on use them on an Onstep board if you go this way.


  • Jaylab likes this

#13 luxo II

luxo II

    Voyager 1

  • *****
  • Posts: 11,560
  • Joined: 13 Jan 2017
  • Loc: Sydney, Australia

Posted 17 September 2023 - 12:33 AM

The mount is a Byers 812.

I'm green with envy.

 

OK those synchro motors and the pear-shaped gearboxes are a standard "industrial timer" form-factor used to drive cams for simple things like 1970's washing machines. Once upon a time there were steppers with gearboxes made in the same form-factor - I came across them in my first job decades ago, though I don't think you will find one now - they are long gone, killed off by digital stuff and the NEMA standard.

 

You have a mount that is well worth keeping and upgrading to stepper control. What I would do is:

 

1. From the number of teeth on the worm wheel, calculate the nominal period of the worm for sidereal tracking. For dec, I'm guessing this has a tangent-arm - measure the radius and the screw-thread teeth and again work out a gearbox ratio to give a similar slew rate in terms of arc-secs/pulse.

 

2. Choose a NEMA stepping motor and gearbox (eg from pololu) to  track at sidereal rate at something like 15 steps/sec, it doesn't have to be an exact match for sidereal - and design whatever metalwork is needed to connect gearbox to worm - this will have to be custom made, but done properly this does not devalue your mount. This will be capable of high-speed slewing and full GOTO.

 

 3. Buy an OnStep kit to drive the motors, assemble and configure that. Some wiring and plugs & sockets involved but all feasible for a beginner if you can use a multimeter to check your connections, and are handy with a soldering-iron or crimp tools.

 

NB I looked at doing this to an old Losmandy G11 that had issues with the Gemini electronics but in the end decided I don't have the time for DIY, and replaced it with a CQ350.


Edited by luxo II, 17 September 2023 - 01:48 AM.

  • Jaylab likes this

#14 TallTanBarbie

TallTanBarbie

    Explorer 1

  • -----
  • topic starter
  • Posts: 85
  • Joined: 20 Mar 2023

Posted 17 September 2023 - 10:17 AM

I'm green with envy.

 

OK those synchro motors and the pear-shaped gearboxes are a standard "industrial timer" form-factor used to drive cams for simple things like 1970's washing machines. Once upon a time there were steppers with gearboxes made in the same form-factor - I came across them in my first job decades ago, though I don't think you will find one now - they are long gone, killed off by digital stuff and the NEMA standard.

 

You have a mount that is well worth keeping and upgrading to stepper control. What I would do is:

 

1. From the number of teeth on the worm wheel, calculate the nominal period of the worm for sidereal tracking. For dec, I'm guessing this has a tangent-arm - measure the radius and the screw-thread teeth and again work out a gearbox ratio to give a similar slew rate in terms of arc-secs/pulse.

 

2. Choose a NEMA stepping motor and gearbox (eg from pololu) to  track at sidereal rate at something like 15 steps/sec, it doesn't have to be an exact match for sidereal - and design whatever metalwork is needed to connect gearbox to worm - this will have to be custom made, but done properly this does not devalue your mount. This will be capable of high-speed slewing and full GOTO.

 

 3. Buy an OnStep kit to drive the motors, assemble and configure that. Some wiring and plugs & sockets involved but all feasible for a beginner if you can use a multimeter to check your connections, and are handy with a soldering-iron or crimp tools.

 

NB I looked at doing this to an old Losmandy G11 that had issues with the Gemini electronics but in the end decided I don't have the time for DIY, and replaced it with a CQ350.

 

 

 

Yep, did the math :)

 

https://www.cloudyni...eq-clock-drive/

 

A regular (i.e. solar) day is 24 hours, or 86400 seconds.  This gives an angular rate of 360 deg / 86400 sec, or 0.0041666666667 deg/sec, which is one degree every 240 seconds, or one degree every four minutes.

 

A sidereal day is approximately 23h 56m 4.0905s, or 86144.0905 seconds.  The angular rate for a sidereal day is 0.0041780746238 deg/sec, or 1.0027379097096 times faster than the regular solar day rate of one degree every four minutes.

 

A clock drive turns the telescope's polar axis at this rate to match the apparent rotation speed of the night sky as the Earth goes around the Sun, which is 1.0027379097096 times faster than a regular clock would turn.

 

Various gear ratios are used to try and match this rate.  One of the most common is to use a 359 tooth worm gear with a 4:1 gear combo attached to a 1 rpm clock motor.  This gives a drive rate of

360 deg * 1/359 * 1/240 sec = 360 deg * 1/359 * 1/4 * 1/60sec = 0.0041782729805 deg/sec, with an accuracy of 4.74756267E-5 compared to the sidereal angular rate of 0.0041780746238 deg/sec.

 

However, large 359 tooth worm gears are getting very difficult to find – and are very expensive if you can find them.

360 tooth worm gears are available, but you need more intermediate gears to get the correct rate with a 1.0027379097096 overdrive.

There are several intermediate gear ratios are available (see attached classic paper from 1934) that give this overdrive rate.

The most accurate of these uses two gear combinations – 51/49 and 79/82.  This gives a drive rate of

360 deg *  1/360 * 79/82 * 51/49  * 1/240 sec = 0.0041780736685 deg/sec, which gives an accuracy of 2.28645573905E-7.

 

I've found good sources for stepper motors and stepper motor gearboxes (not cheap either), so if for example I used a 60:1 gearbox on a NEMA 23 stepper motor with 200 steps per revolution, I'd have

360 deg *  1/360 * 79/82 * 51/49  * 1/60 * 1/4 sec = 360 deg *  1/360 * 79/82 * 51/49  * 1/60  * 1/200 steps * 50 steps per second = 0.0041780736685 deg/sec, but I still need sources for small 79, 82, 51, and 49 tooth gears.

 

 

Thanks to user Mercury-Atlas, found some of those old Hurst stepper motors that come in the same housing ( https://www.electron...er-12vdc-3-watt ).  I ordered a couple of them, so we'll see if they fit.

 

 

What I need now is something to hook those surplus motors to see if they actually run.  I can't find any info about that specific motor, did find some general information on the old Hurst web page ( https://www.hurst-mo...sabsgeared.html ), so all I know now is that it has a 20:1 gear reduction and maybe a 7.5° step (48 steps per rev). 

 

 

My guess from looking at the intermediate gears inside the mount casing is that it's a 2:1 ratio, so that means that the 1/240 rate to turn the 359 worm gear would be 1/2 * 1/10 gear reduction * 1/12 sec = 1/240 if the motor is 5 RPM. 

 

 

If these old surplus steppers are geared down to 20:1 as advertised, then the setup would be 1/2 * 1/20 gear reduction * 1/6 sec = 1/240, so I'd have to run the motor at 10 RPM, or eight steps per second ( 0.125 step /sec, one step every 125 milliseconds) for the sidereal rate, sixteen steps a second for 2x, twenty-four steps per second for 3X speed, etc.



#15 BGRE

BGRE

    Fly Me to the Moon

  • *****
  • Posts: 5,191
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 17 September 2023 - 06:36 PM

2 arcsec per step is a bit coarse.

Something like .1 arcsec/step or finer is desirable at least for imaging.

Microstepping is one approach.



#16 TallTanBarbie

TallTanBarbie

    Explorer 1

  • -----
  • topic starter
  • Posts: 85
  • Joined: 20 Mar 2023

Posted 17 September 2023 - 08:21 PM

2 arcsec per step is a bit coarse.

Something like .1 arcsec/step or finer is desirable at least for imaging.

Microstepping is one approach.

I'd thought about using contemporary NEMA 23 stepper motors, there's just not a lot of room to mount the motor and a reduction gearbox, and it still has to fit on the trapeze that flips down to engage the worm gear. 

 

Don't even know if these old surplus Hurst steppers that I ordered even work yet, we shall see.

 

That's still the backup plan if none of the other options work.

 

 
byers gear 2
byers gear 1


#17 gregj888

gregj888

    Gemini

  • -----
  • Posts: 3,141
  • Joined: 26 Mar 2006
  • Loc: Oregon

Posted 19 September 2023 - 11:15 PM

I sued the pear shaped stepped with a Mel Bartel board on my houghton.  I upgraded to a Sidral tech servo system but the steppers did work.

 

I thought there were encoder and limit switch options for Onstep.  There is a single limit and no encoder input which may be enough.  That's even available on the mini boards.  If you want to do the minimal amount of soldering, use one of the #D printer controller options. 

 

Attached is my Meade SCT using a stepper like this one running off of 12 volts--

 

https://www.ebay.com...:Bk9SR5rE04zWYg

 

IMG_3170.jpg


Edited by gregj888, 19 September 2023 - 11:16 PM.

  • TallTanBarbie likes this

#18 22SQM

22SQM

    Explorer 1

  • -----
  • Posts: 85
  • Joined: 27 Aug 2023
  • Loc: Rio Rancho, NM

Posted 29 October 2023 - 05:02 PM

I used a $100 VFD from China to control my Bodine 3-phase AC motor frequency.
The VFD is Programmable with Function values and can accept the ST4 signals from the guide camera (and also the hand controller for slewing).
Its a bit overkill, but $100 isn’t that much.
https://www.cloudyni...-1693276750.jpg

Edited by 22SQM, 29 October 2023 - 06:34 PM.


#19 don clement

don clement

    Vendor (Clement Focuser)

  • -----
  • Vendors
  • Posts: 2,617
  • Joined: 02 Feb 2011
  • Loc: Running Springs, California

Posted 30 October 2023 - 06:19 PM

One is not going to make a goto drive using an 812  equatorial mount that uses a  section of a worm to drive the RA axis and must be reset every two hours. So why worry about changing speeds very much on  the AC motor? 




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics