slowing down a 9v Quartz controlled motor

Discussion in 'The Projects Forum' started by Davidb_54, Dec 7, 2011.

  1. Davidb_54

    Thread Starter New Member

    Dec 7, 2011
    7
    0
    Hi everyone,
    I am an ageing astronomer and in NO WAY not an electronics engineer, but I have an electronic problem.
    I am using a commercially supplied Celestron 9v DC motor and high reduction gearbox system which gives a speed at the output shaft of 1 rev per 11 mins, with good high torque.
    The motor is described by Celestron as quartz controlled, the speed control is very accurate, adjusted by a small rotary spindle ( potentiometer thing ?), mounted on a board with a bunch of other resistors and stuff and what looks like a square black bug with 8 legs (!!)
    BUT the output runs just too fast for my application. At the slowest end of the spindle setting it is still about 10% too fast for my needs.
    Is there a simple way to reduce the speed output, by say adding another resistor to the control circuit perhaps ?
    Gears etc are too clumsy and I am Sooo ... close to getting the speed I need from this great little unit.

    Any suggestions would be very gratefully recieved.

    thanks in advance for yout time

    davidB
     
  2. wayneh

    Expert

    Sep 9, 2010
    12,090
    3,027
    It'd be great if you could post a schematic. Assuming you don't have that, a clear photo of the circuit would help, or a link to the manufacturer with a model number. You might contact the manufacturer, by the way. I've had success getting repairs on devices years out of warranty. Just depends on the company.
     
    Davidb_54 likes this.
  3. Davidb_54

    Thread Starter New Member

    Dec 7, 2011
    7
    0
    Thanks for your time.
    as you suspected, I do not have a schemetic ,but will try to post decent photos of the circuit board etc when I get home tonight
    thanks again
    davidB
     
  4. MrChips

    Moderator

    Oct 2, 2009
    12,420
    3,355
    Are you using this in a telescope tracking system?
    The reason I ask is because if the telescope mount is aligned properly it should track correctly without any modifications.
     
  5. Davidb_54

    Thread Starter New Member

    Dec 7, 2011
    7
    0
    Hi Again
    here are some images of the commercial unit I want to slow by about 10%
    the two switches ate on/off and forward/reverse
    RED LED power on light and spindle pot
    gearbox gives a huge ca 1000x reduction , but I still need to slow down about 10% more than the adjuster knob range gives me.
    thanks again

    davidB
     
  6. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    In one of the photos, there is what looks like a battery clip for a 9v battery. Is that the sole souce of power for the circuit board and motor? And does the motor speed vary as the battery runs down?

    Please read the letters and numbers that are on the top of the eight pin IC, and post them. Also on the transistor (black three legged device.)
     
    Davidb_54 likes this.
  7. Davidb_54

    Thread Starter New Member

    Dec 7, 2011
    7
    0
    Hi
    yes the sole power to both motor and board is from a single 9vPP3
    the motor does not slow down noticeably when running under load and battery is low, rather finally just stalls under load and stops, with the power light still illuminated
    10uf & -25 volts on the round thing with a white top
    so331 on the other black cylinder near the 8 pin IC, the transistor ?
    and on the IC
    CHM
    CM358N
    KQF974

    thanks to my daughters amazing eyes for the data, as I can't see ANYTHING on these components at all !!

    thanks again for your time

    davidB
     
    Last edited: Dec 8, 2011
  8. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    It doesn't seem to be "quartz controlled". I can't see a crystal. (maybe it's hidden on the back?)

    In order to have an idea of what the schematic looks like please take a picture (with good resolution like the others) of the back of the circuit board. All the traces should be visible on the picture.
     
  9. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    The 8pin IC is a LM358 type dual-opamp.
    The "331" 3 pin package is a mystery! Are there any more numbers on it?

    My GUESS, as the motor has only 2 wires, is that it is a low current DC motor. Then the LM358 is probably configured as a precision voltage source, so the motor speed is set in open loop fashion by providing the motor with a precise unchanging DC voltage.

    I does not look to be a frequency driven system to me, based on the fact there are almost no caps (it would need timing caps) and the motor is very unlikely to be a 2-pin AC motor I think.

    If you had equipment (like a 'scope), you could run the device and measure the waveform on the 2 motor terminals. Even without a scope you could use a typical multimeter and measure DC volts on the motor terminals when it is running, and also measure AC volts. That information would help a lot. :)

    Other helpful information would be to measure DC volts on the middle pin of the adjuster pot (referenced to 9v battery -ve terminal). I expect you will find as you adjust the pot the DC volts on the middle pin go up and down, but other wise is a stable DC voltage, which might indicate that it is a linear voltage regulator.

    If that is the case it will be easy for you to modify, but the crude nature of a DC regulated motor controller is that it will require a weight balanced load (balanced telescope).
     
  10. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    I think you are right. A marketing guy (I was one) could justify (in his own mind) calling it quartz controlled because of the silicon based semiconductors.

    In the photo's, it looks like the transistor's flat side (where the ID numbers would be) is too close to the switch to be legible, but maybe the OP can look again?
     
    Last edited: Dec 9, 2011
  11. wayneh

    Expert

    Sep 9, 2010
    12,090
    3,027
    I wonder if this isn't the key. It seems odd that the controller has failed without failing catastrophically, and it makes me wonder if there isn't something else afoot, like an unbalance causing excessive mechanical load (motor speed over corrects?) or a negative load, making the motor spin faster than it wants. Or a misalignment as previously suggested. I think it would be wise to rule out mechanical issues before tearing into the controller too much.

    On the upside, that controller should be a snap to work on if need be.
     
  12. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
  13. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Yes the LM331 is, but it is not a 3 pin TO92 package, LM331 is only 8pin package.

    My guess of the circuit is that the glass diode is a voltage reference, possibly fed by one side of the op amp as a fixed current source.

    The other side of the opamp would be the closed loop voltage control, comparing motor voltage to the pot setpoint voltage. The 3 pin device is probably a garden variety 500mA PNP or FET that controls motor voltage (from the output of the opamp).

    The direction switch is a DPDT reversing switch (as it is probably a DC motor).

    The whole thing is just a precise adjustable voltage supply, that sets the voltage to the DC motor and it relies on a weight-balanced load and the 100:1 gearbox to reduce any load induced speed variation. That pretty much matches every component.
     
    Last edited: Dec 9, 2011
    Davidb_54 likes this.
  14. benemorius

    New Member

    Dec 29, 2011
    1
    2
    I appear to have exactly the same motor drive as you. Assuming I'm allowed to attach something with a post count of 0, I'm attaching a schematic I made of mine. Be aware that I haven't spent any time checking it for accuracy.

    I found this thread while searching for someone else with the same problem as me. The trouble with my unit is that it won't run fast enough at colder (<15C) temperatures, and only barely runs fast enough at 23C. Furthermore, the upper 50% of the pot adjustment is useless. That is, from 50% to 100% the motor speed remains constant rather than increasing. I took it apart and made a schematic in the process of looking for a solution.

    The solution I currently have in mind will involve replacing the opamp with a rail to rail equivalent. The output of the opamp is clipping at around 6v, which is why the motor won't go any faster past 50% on the pot. That's 8.5v battery input minus 1v reverse current protection diode drop (the glass one; the voltage reference is actually the LED) minus 1.5v opamp swing limit. I haven't replaced the opamp yet and I don't know if further modification will be required to get full speed. Actually I'm not even sure yet which part is responsible for the severe temperature drift. Clearly this is either a design flaw or someone is populating units with incorrect parts. I'm somewhat puzzled by the fact that I haven't found numerous people with the same complaint as me, and certainly quite surprised to see someone with almost the exact opposite complaint.

    Back on topic now - the schematic should be good enough to facilitate finding a solution for your fast motor. It is most definitely NOT quartz controlled. In fact, it's not even very precise. In any case, it will be easy to slow down. It would appear that stacking an additional 9.1k (or 10k, or anything close) in parallel with the existing 9.1k resistor (R5 - directly over the 8 pin IC) OR replacing the 1k (R6 - directly next to the 9.1k) with a 2k (or close) should cut the speed in half across the board by reducing the opamp gain from about 10 to about 5. That might put your target speed somewhere around the middle of the adjustment range. Then again, it might not.

    PS: I simply must ask for the sake of curiosity - are you actually using this motor on a telescope? Is it the correct motor for the telescope? Can you discern any change in speed while turning the pot in the upper 50% of its range? Was the 1 rotation per 11 minutes your own measurement? If so, at what speed setting on the pot did you check it? Mine runs at 1 rotation in just under 10 minutes on the fastest setting; 24 on the slowest. Sorry for all the questions. My mind just isn't one that handles an unsolved mystery very well. I'm curious to know the story behind these and how they can be selling them like this.



    keywords: celestron astromaster logic drive precision dc motor drive 93514 too slow too fast
     
    Davidb_54 and SgtWookie like this.
  15. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    LEDs are quite temperature dependent as voltage references. Also the motor/gearbox characteristics (grease getting thicker etc) will cause the motor to run slower at colder temperatures for the same regulated motor drive voltage.

    Could you change the motor for a stepper motor? I recently did a open-source stepper motor xtal-locked telescope driver, adjustable to the nanosecond, using the Linistepper kit and free PIC firmware. A xtal-locked stepper motor gives you professional performance telescope tracking.
     
  16. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Benemorius,
    Thanks a great deal for your post, and particularly for taking the time to make the schematic and post it.

    I'll suggest that your difficulties in cooler weather are due to multiple issues:

    1) The output of batteries decreases as the temperature goes down.

    2) The diode D1 seems unnecessary, and it will drop approximately 0.8v when the motor is running. The colder D1 is, the higher its forward voltage will be for a given current.

    3) The colder the opamp is, the greater the gap between Vcc and the maximum I/O voltage.

    4) The base-emitter drop on Q1 will increase as the temp goes down.

    5) Incorrect lubricant used.

    So to recap:
    Battery output down
    D1's 0.8v drop increase
    U1's I/O limit increasing from Vcc-~1.2v to Vcc-~1.6v, and the current limit of U1's output when near the limits.
    Q1's BE Vf; ~0.8v; increases when cold.
    Lubricants should be synthetic so the viscosity does not change over temperature.

    Suggestions:
    1) Use an industrial 9v battery; they have 7 cells internally rather than 6 cells. A standard 9v pp3 battery puts out under 9v when loaded even when new. Industrial 9v cells put out ~10.4v. Just changing battery types may take care of your situation.

    2) Either replace D1 with a piece of wire, or a low-Vf diode like a 1N5817 Schottky diode.

    3) Replace U1 with an RRIO opamp.

    4) It may help to add a 2k pull-up resistor from the base of Q1 to the 9v source.

    5) Remove all traces of lubricants that are not synthetic using 99% isopropyl alcohol, allow to dry, and apply synthetic lubricant of the proper viscosity.
     
  17. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I didn't realize it was taking me so long to reply; Roman's post wasn't there.

    That's what I also initially jumped on - but in this case, it's good that the Vf of the LED increases, as the output of the 1st opamp will be higher, and that will help to compensate for the mechanics being more sluggish (particularly the gearbox if the lube is not synthetic).
     
  18. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Thanks for the correction! I knew LEDs were temperature dependent for Vf but wasn't sure which way. :)

    Re the regulator circuit, it would be better if using a PNP as the output transistor.

    It might be as easy as swapping the NPN to a PNP, then reversing pins 5 and 6 on the opamp (to reverse opamp input polarity?). That would give an improved max motor voltage up to about 8v?

    But still, changing to a stepper motor and free PIC firmware (which is already done) will give adjustment for any gear ratio and allow pro quality telescope tracking. That's definitely worth looking at.
     
  19. DickCappels

    Moderator

    Aug 21, 2008
    2,647
    632
    As this appears to me a simple DC motor, it seems that the simplest and safest way to make it go at 90% of the speed its now going, is to put a 68k resistor in parallel with R5, reducing the gain of the first opamp from 10 to 9. Am I missing something?
     
    Davidb_54 likes this.
  20. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Hi Dick, I think the problem is the opamp is limited to a positive output of V+ -1.5v or so, so with a battery reading 8.5v and another 0.7v lost on the diode, 1.5 lost on opamp and then 0.6v lost due to using a NPN emitter follower it tops out at max of about 5.7v at the motor.
     
Loading...