Jump to content


Photo

Old non-Goto CG-5 hand controller died

  • Please log in to reply
39 replies to this topic

#26 DavidNealMinnick

DavidNealMinnick

    Ranger 4

  • *****
  • Posts: 360
  • Joined: 06 Mar 2006

Posted 15 February 2011 - 06:54 AM

I ordered a spare controller directly from Celestron a few years ago anticipating just this scenario. I think the price was $20. Perhaps they still have them?

#27 TrackballJerry

TrackballJerry

    Explorer 1

  • -----
  • Posts: 54
  • Joined: 14 Apr 2010
  • Loc: Eugene, Oregon, USA

Posted 16 February 2011 - 12:51 AM

you might also want to add RA guiding to that :-)...unless of course your needs are purely visual.


At this point visual is all I do. I used to do photography, but I got tired of always fussing with the gear. Nowadays I enjoy the sky when I'm out and fuss with the gear on cloudy days.

I ordered a spare controller directly from Celestron a few years ago anticipating just this scenario. I think the price was $20. Perhaps they still have them?


Who did you talk to? I wrote to Celestron tech support when the controller first died, but they just said it was discontinued and sent me elsewhere. Maybe I should have gone to the sales department? Heck, for $20 I'd get one even though I've already got the parts coming for the single-axis mod.

Orly, a question: On your website describing the controller mod you say that the Atmel micro-controller chip is good up to 24 MHz. If I put a 5 MHz crystal in and then hit the 8x slew button, am I going to over-drive the chip, or will it still be running at 5 MHz no matter what the slew rate is?

Jerry

#28 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5361
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 16 February 2011 - 12:59 AM

Jerry,

No you won't overdrive the chip. The chip always runs at the crystal clock, which in this case would be 5 MHz.

The 8X button just tells the chip to output the stepper pulses 8X faster, but the stepper pulse rate never exceeds about 300 pulses per second, that's the maximum that the stepper motor can take.

#29 TrackballJerry

TrackballJerry

    Explorer 1

  • -----
  • Posts: 54
  • Joined: 14 Apr 2010
  • Loc: Eugene, Oregon, USA

Posted 17 February 2011 - 09:44 PM

Cool! The crystals arrived in today's mail, so as soon as the controllers show up, I'm set.

These crystals are tiny! They're little cylinders about 1/8 inch wide and 1/2 inch long. Hope I got the right ones. They were the only 5.12 MHz crystals DigiKey had in a metal canister shape, so that's what I got. There'll be no trouble fitting them in the case, that's for sure.

Jerry

#30 TrackballJerry

TrackballJerry

    Explorer 1

  • -----
  • Posts: 54
  • Joined: 14 Apr 2010
  • Loc: Eugene, Oregon, USA

Posted 22 February 2011 - 09:55 PM

I've hit a snag. The replacement hand controllers arrived today (I bought two, 'cause they were so cheap), but they drive the motor too fast, not too slow. My mount is apparently a different design than Orly's. My mount has the motor driving the worm gear directly, while this setup is designed to drive a worm gear with what's apparently a 5:1 reduction gear between motor and worm. So instead of spinning too slowly, as Orly's did, this one is spinning 5 times too fast. (It spins at 30 rph, while I need 6 rph.)

I took one of the controllers apart to see if I could be lucky enough to have a 25 MHz crystal in there (which I could then replace with my 5 MHz crystal and be good to go), and received a second rude surprise: the labels on the crystal and on both ICs have been scratched out. Somebody deliberately wiped them out so you can't tell what values they are! That's pretty sleazy.

I'll proabably swap the crystal just to see what happens (curiosity killed the cat), but I'm not hopeful.

I can probably rig a 5:1 reduction gear between the motor and the worm. I'll have to see what I've got for gears and room to put them in. Not an elegant solution, but it'll probably work.

Jerry

#31 TrackballJerry

TrackballJerry

    Explorer 1

  • -----
  • Posts: 54
  • Joined: 14 Apr 2010
  • Loc: Eugene, Oregon, USA

Posted 22 February 2011 - 10:19 PM

I just swapped the crystal and the motor runs 1.4 times faster, so they must be using a 3.58 MHz crystal like they do in the other controllers.

I checked DigiKey to see if they sell 0.716 MHz crystals, which is what I would need to slow this down by a factor of 5, but that's apparently slower than crystals go. They didn't offer anything below 1 MHz, anyway.

Hmm. Looks like I'll need gearing, unless somebody here has a better idea.

Jerry

#32 TrackballJerry

TrackballJerry

    Explorer 1

  • -----
  • Posts: 54
  • Joined: 14 Apr 2010
  • Loc: Eugene, Oregon, USA

Posted 24 February 2011 - 06:00 PM

I looked at the Ebay ad again for the controllers I bought, and the ad says the motors turn at 14.4 minutes/revolution, so neither Orly nor I were barking up the wrong tree. I was just sent the wrong part, or the description in the ad was wrong. I've contacted the supplier to see what we can do about it.

Jerry

#33 TrackballJerry

TrackballJerry

    Explorer 1

  • -----
  • Posts: 54
  • Joined: 14 Apr 2010
  • Loc: Eugene, Oregon, USA

Posted 14 March 2011 - 08:11 PM

I received the Orion TrueTrack drive today, the one that I was hoping would be a plug-in replacement for the Celestron drive that died. It wasn't quite, but it was close enough that I could steal parts from the old drive to make the new one work.

I didn't trust the motors on the old setup, thinking it was possible that a short in one of them could have burned out the old controller, so I wanted to swap out the motors as well as the controller. The mounting brackets were just different enough that I couldn't do that directly. I had to swap the gearbox housing on the RA motors so the bolt holes would line up properly, and I had to make a completely new bracket to hold the RA power plug. That was it, though. Once I made those two fixes, it bolted into place and runs just like the old one.

The plug for the dec motor is different from the old one, too, so if I'd wanted to use the new controller on the old motors, I would have had to change the plug for the dec motor.

I haven't yet tested to make sure the tracking rate is the same, but the gear ratios look the same and the motor speed sounds the same as the old one, so I'm betting it is.

I opened up the new hand controller and it's the very same inside as the old one, right down to the 40-pin chip for a brain. I'm very tempted to swap the old chip into the new controller to see if it's still functional. If it is, it would be worth the trouble to replace the other parts in the old controller one by one until I got it working again, if only to have a spare when the new one eventually dies. But I'm wondering how wise it would be to plug a possibly burned-out chip into the new controller. What are the odds that it could wipe out other components? I'm pretty sure I don't want to go the other way (putting the new chip in the old controller). What do you think?

Jerry

#34 TrackballJerry

TrackballJerry

    Explorer 1

  • -----
  • Posts: 54
  • Joined: 14 Apr 2010
  • Loc: Eugene, Oregon, USA

Posted 23 April 2011 - 09:37 PM

I finally got a chance to take the mount out for a night of observing, and I'm happy to report that the tracking rate is spot on. So it's official: you can use the Orion TrueTrack drive on a Celestron CG-5, with just a little bit of modification to the motor mounts.

I found a more stable power supply for it, too. For my old controller I was using a 6-volt supply that actually put out more like 6.8 - 6.5 volts depending on load, and I was afraid that might have been what burned it out, so I found a 5-volt supply that actually puts out 5.5 volts. It drops to 5.4 under a 500 milliamp load, which is about what the motors draw at highest drain. (Surprisingly, that's at 2x slew speed, not 8x.) The controller was designed to run off 4 D-cell batteries, so it seemed likely that it would be happy with 5.5 volts (about what a set of batteries would put out when they're partially discharged), and it did indeed run without complaint on 5.5. And I'm much less worried now about burning out the new controller.

Jerry

#35 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5361
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 24 April 2011 - 03:15 AM

Congrats Jerry. The Vixen DD-1 is also susceptible to burning out, so my solution was to use an overly-small power supply that only puts out 300mA. The power supply gets hot though.

#36 TrackballJerry

TrackballJerry

    Explorer 1

  • -----
  • Posts: 54
  • Joined: 14 Apr 2010
  • Loc: Eugene, Oregon, USA

Posted 24 April 2011 - 01:53 PM

Hi, Orly,

What do you think of the idea of me taking the 40-pin ROM out of my old controller and testing it in my new one? If the ROM is good, I'd happily replace other components until I got the old controller working again (always good to have a backup, and at $150 for a new one, $20 or so on diodes and transistors to fix the old one seems like a pretty good deal), but I don't know how else to test the ROM. Is there much danger that putting a bad ROM into the new controller could damage the new controller?

I'm willing to take *some* risk, just not a stupid risk, and I don't know enough about electronics to know whether this is a stupid risk.

Jerry

#37 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5361
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 24 April 2011 - 08:33 PM

Jerry, the big chip is not a ROM. It's an 8051 clone. I assume they all are pin-compatible with each other (since they are all Intel 8051 clones). Putting the old CPU in the new controller should be fairly risk-free, these old CPU's are quite robust.

In your old controller it's likely that the power drivers are busted, not the CPU.

#38 TrackballJerry

TrackballJerry

    Explorer 1

  • -----
  • Posts: 54
  • Joined: 14 Apr 2010
  • Loc: Eugene, Oregon, USA

Posted 25 April 2011 - 10:01 PM

Jerry, the big chip is not a ROM. It's an 8051 clone. I assume they all are pin-compatible with each other (since they are all Intel 8051 clones). Putting the old CPU in the new controller should be fairly risk-free, these old CPU's are quite robust.

In your old controller it's likely that the power drivers are busted, not the CPU.


I swapped the chips, and sure enough, the old one works just fine in the new controller. Yay!

So when you say that the power drivers are busted, are you talking about the thicket of transistors and diodes just before the wires going to the motors? One of my VOMs has a transistor and diode tester, so I should theoretically be able to spot a bad one. Lots of unsoldering, but worth it if I can fix the controller.

Jerry

#39 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5361
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 26 April 2011 - 03:56 AM

Yup most likely one of the transistors is busted.

Ideally one should have snubber diodes (fast schottky types) for motor drivers, but schottky's are expensive so most folks cheap out and use the usual 1N4001, which is not fast enough that sometimes motor spikes get through and fry the transistor that they are supposed to be protecting.

I know it's one of those "not worth your time" deals but replacing all the transistors, and all of the diodes (I think there are only 4 of each -- oops, 4 of each, per motor) is probably the most straightforward bet. If you're feeling lavish, replace the transistors with metal ones. You'll have to check if they are PNP or NPN though (your tester will help) if the Chinese have scrubbed off the markings on them... as for the diodes, those are easy to test but if you replace them with 1N4007 (400V rated) you'll never risk burning them out. Heh.

#40 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5361
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 26 April 2011 - 03:59 AM

One thing you can try is.. try measuring the voltage at each wire going to the motors while the controller is on. If that particular transistor/diode pair is OK, you should get a pulsing signal. Look for the wire which is dead, Jim.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics