I worry a lot.
It seems there are a lot of folks that have trouble with hooking up their guider correctly, things not working right, and even damage to the mount and/or camera…. So I worry. After I got my Losmandy mount with the Gemini controller I reviewed the manual carefully and read lots of conflicting posts on the subject trying to be absolutely sure that I wouldn’t damage anything (in my mind, downtime + repair cost = BAD). What I read didn’t make me feel secure. Although I’m pretty sure an SBIG ST series camera guide port can be hooked up directly to a modern Gemini without a problem, I’m not absolutely, positively, completely sure. The issue has to do with forming ground loops between the power supplies to the camera, laptop, and mount. It seems like there’s a possibility that the signal grounds of the guider could return via different paths and form a ground loop. Not good for guiding, an potentially dangerous. And as I think is abundantly clear, I worry a lot.
So, I decided to optically isolate the camera guide port from the Gemini. I found that I could buy a optically isolated cable ready made for $70 from Tom Hilton (www.arizonaskys.com). However, being cheap, I decided to build one myself. I’m not actually advocating building one since it’s a bit of work, and probably only saved about $50, but still, I like these projects. Plus, going through the process made me really understand how the guide circuits work.
The first step was to find how complicated a optocoupler circuit would be. I found a good diagram here: SBIG_opto_isolator.jpg (I also found an almost identical circuit from Mike Dodd posted on either the Gemini or Losmandy user groups on Yahoo. If these links become broken in the future, send me a note (altuttle <at> alberts-astro.com) I may be able to scrounge up the circuit diagram. Anyway, although the circuit might look complicated at first, with a little study you see how easy it is.
Power to run the opto-isolating IC chips is provided by the camera. For each direction, the power is supplied to pin 1 of the chip and returned to the camera direction pin via a 150 ohm resistor. An optional LED is also in the circuit to show when that direction is signaled (very cool feature). What happens when the camera “signals” a direction, the appropriate pin on the camera is shunted to ground. So you can see what happens is that completes the circuit which would energize the chip and light the LED. It’s that simple. Then the output of each chip is connected such that when energized it will shunt the corresponding signal pin on the Gemini to the gemini ground. Thus there is NO direct ground or voltage connection between the SBIG and the Gemini. Easy!
So, to build it, I got a 6 conductor phone extension cable from Radio Shack and I cut a length off one end. I already had the LEDs, resistors, little circuit board, and a male DB-9 connector, but they can be had from Rat Shack as well (total cost should be less than $15). The optocoupler ICs I bought from Digikey, This cost about $3, but then there’s the $7 worth of shipping (bleh). For the box to put it in, I used a case from a wall wart power supply (might have been a phone charger or something) that I picked up at Goodwill for a dollar or two.
This is what it looks like:
And this is how it looks on the camera:
I used it for the first time a few weeks ago. It works great, and having the LEDs indicate each time a signal is sent is really cool. I can tell if there’s a problem with my calibration, or if my DEC is pushing back and forth, or whatever…. This actually happened during my first use. I wasn’t familiar with how the SBIG software (CCDOps) calibrated and that gave me trouble. What was happening is that after CCDOps reported a successful calibration and I started guiding, my guide graph looked lousy, specifically in one direction. Just by looking at my box it was obvious there were no pulses being sent to one of the directions. A little checking showed me that that the software had turned off guiding in that direction. Turns out it happened because the calibration star went off-screen during the calibration process. Okay… re-do the cal process making sure the star stays on-screen and all was well. Bottom line, I like the indicators.
Good deal, the guider is totally isolated and works fine. One less thing to worry about.