ETX-125 DEC DRIVE MODIFICATION Last updated: 16 June 2009

Sent:	Friday, April 17, 2009 15:10:21
From:	John Farrant (johnfarrant@gmail.com)
A follow up to my DEC board problems.
 
This project started when my dec drive board on my 5 year old ETX 125
died on me.

The symptoms were that when an attempt was made to slew up or down, the
power light would dim. There was little or no response from the motor. I
checked the supply rail while this was happening. The 12 volts had
dropped to about 4.5!
 
I had 3 choices:
 
1. Return the scope to Meade.
2. Repair the board myself.
3. Use what was still working.
 
Option 1 was out of the question.
 
I also realised that repairing the board myself was going to be
difficult, even assuming that I could obtain the two output fets, a
comparator and a reprogrammed pic.

The odd thing about this is that I don't understand why the dec board
died. Two output fets, the pwm chip plus part of the pic? I have a
theory about the longer power leads supplying the dec board picking up
spikes, but it's only a theory.

Thankfully the RA board was still working ok.

I first removed the short on the dec board by lifting pins 1 and 3 from
the motor drive chip. This at least allowed the Autostar 497 to
recognise that the dec board was connected and allowed a succesfull
initialise. Unfortunately, when the software needed to communicate with
dec board, ie when tracking, a motor fault message appeared. This only
occured when the scope was in the alt/az mode. In polar mode, as only
the RA motor is driven, tracking was ok.

I decided that my only option was to operate the dec motor from a
separate power source. I rigged up four 1.2 volt rechargeables, together
with a two pole toggle switch. This gave me 2.4 volts with switchable
polarity - motor up and down. The batteries were fitted in the normal
battery compartment which was rewired to suit the new power requirements
of the dec motor. I found another switch which allowed me to supply the
full 4.8 volts to the motor. This gave me two slew speeds - 0.5 deg/sec
and approx 1.5 deg/sec.

Initially, I had the new dec motor power lead passing through a small
hole drilled in the side of the plastic fork arm near the motor. This
looked untidy. I thought about using the unused dec power leads -
red/black - to supply current to the motor. However, when I unplugged
the four pin dec plug, the RA motor wouldn't respond to speed commands.
It just ran away at max speed. On a hunch, I reconnected only the blue
and yellow leads leaving the red/black disconnected. Amazingly, no
runaway! The HBX booted up OK! How, with no power supplied to the board
was this happening? Another thing I found was that when the normal
bootup had completed, I could unclip the blue/yellow leads and the
Autostar was quite happy. I tried terminating the blue/yellow with a 250
kohm resistor - equivalent to the input R of the dec board - but it
didn't work. This was a pity as I had thought of another use for these
two leads.

With my new controller operating ok, and the dec motor power leads 
running neatly through the centre RA bolt, I turned my attention to my
GOTO options.

In polar mode, GOTO is impossible as the dec motor needs to run. I had
half a chance in alt/az if I operated the scope in terrestrial mode.

All I needed to do was, 1. Write an Equatorial to Horizon coordinate
conversion program, 2. Drive the RA motor to the correct AZ position,
and 3. Stop the drive when the desired AZ position was reached. Then use
my gizmo to move the scope to the correct altitude of the desired object
- using the ETX's setting circle.

Together with  Peter Duffet-Smith's excellent book, Practical Astronomy
with your calculator, my BlueTooth PDA and a bit of C programming, I
managed the first part of the problem without too much difficulty.

Thanks to the knowledge aquired from this site, I used the LX200
commands to slew, at least in AZ, to any object above the horizon.

My final hurdle was how to get the scope to stop slewing when it reached
the correct azimuth? This proved supprisingly easy.

Simply check the scopes AZ position twice every two seconds or so and
compare the two results. If they are equal, then the final position has
been reached. All that's required then is to hit the mode key before the
motor fault message appears. This message is followed by a motor test
and a reboot. There's about a twelve second delay before this message
comes up after a slew has terminated.

As the AZ display doesn't update untill after the slew has completed,
all of this has to be done using the LX200 command set. It cannot be
done manually without listening to the motors. Not very practical.

This is part of the C code which checks when the azimuth slew has
finished...

Note. ret is the handle to my PDA Bluetooth port.

If anyone wants the complete C code for slewing (in azimuth only of
course) to selected Messier objects, give me a shout.
 
readchar(string item)
{
   reply=item;
   do
    {
      reply=reply+cc;
    }
    while(cc!="")
  sleep(200);
return();
}
 
//check if slew almost
//completed-
//every 2 seconds ..
//Note. All timings are fairly critical
do
{
 sleep(400);
 writebytes(ret,"#:GZ#"); // check AZ position
 sleep(300);
 readchar("");
 az1=strleft(reply,3);
 sleep(2000);
 writebytes(ret,"#:GZ#"); // and again ..
 sleep(300);
 readchar("");
 az2=strleft(reply,3);
 puts(az1+", ");
 putsl(az2);
 if(az1==az2) // if two identical readings, final AZ position reached!
  break;
}
while(az1!=az2)
 sleep(400);
//kill motor fault
//message with EK9 .. equivalent to pressing the Mode key ..
 writebytes(ret,"#:EK9#");
 sleep(400);
//print report ..
 putsl("Slew completed ..");
  putsl("Alt: "+bluealt+" Az: "+blueaz);
getc();

On the night of April 13th, using my Bluetooth PDA and a RS232/Bluetooth
adaptor, I logged:
M44, M48, Regulus, Procyon, Denebola, Alphard, Algiea, Saturn, M36, M93,
M50, M47, M1, Castor, Betelgeuse, thCancer.

Castor and M1 were in the field of the 26mm eyepiece, and the rest were
just outside - but well within the finder.

All without the DEC board powered up!

I've ordered a Digital Inclinometer with a 0.1 degree resolution - about
25. This should inprove my success rate. Also it will eliminate base
slope errors. Any error will be in azimuth only. The altitude must be
correct - at least to +/- 6 minutes or so. Plus I don't need to enter
the date/ time etc into the Autostar. The PDA provides it.

I'm now a happy chappie. My ETX may not be as it was, but at least I can
continue my favourite hobby.
 
Clear skies,
 
John

Sent:	Sunday, June 14, 2009 14:00:37 
From:	John Farrant (johnfarrant@gmail.com)
ETX-125 DEC DRIVE MODIFICATION

Latest update ..

You may remember (04/20/09) that the DEC board on my ETX-125 suffered a
serious malfunction. Initially I suspected the PIC together with the
power MOSFETS used to drive the motor. The saga continues ...

At a local second hand fair I picked up what I assume is a DS2070. This
I obtained for the princely sum of 20 euros (28 dollars). It had
suffered considerable water damage, although the OTA seemed in good
condition. The mount on the other hand was a different story. The RA
bearing had completely seized. No amount of twisting or turning would
shift it. After a lot of huffing and puffing I managed to remove the
base to be greeted by a considerable amount of water damage. The
internals were a mess.

I removed the RA and DEC motors together with the single printed circuit
board containing the drive electronics. I powered up the board and found
the DEC drive was ok but the RA was dead. The LED in the RA encoder was
open circuit. Undaunted I considered my options. Could I use anything on
the DS board to repair my ETX?

The DEC motor driver used in the ETX-125 is made up of four MOSFETS - 2
N-Channel and 2 P-Channel each in a single dual package. On my DEC board
the Dual P-Channel chip had blown. The P-Channel IC on the ETX-125 DEC
board is a single 4947ADY. This is the IC where the source of each
MOSFET is connected directly to the 12 volt supply. The drains by the
way are linked to the motor leads. Thanks to Dick for this information.

The P-Channel MOSFET fitted to the DS2070 is an NDS9953A. This has
similar specs to the 4947ADY and I figured that it should be a fair
replacement - not exact, but fair.

Anyway, to cut a long story short, I removed the NDS9953A from the DS
board and soldered it in the place of the 4947ADY on the ETX dec board.
This took over 3 hours! Not recommended for the faint hearted as 
MOSFETS have a habit of disappearing in a cloud of static. I was lucky
and only lost a small section of copper track which I had to link.

Did it work? Yes, although the DEC drive is not quite as smooth as it
was. At slew speeds it's a little jerky at the initial spin-up. Seems to
settle down though. Tracking is ok.

I'm hoping that the replacement MOSFET survives.

I'll keep you posted either way.

Kind Regards,
John Farrant

Go back to the Tech Tips page.

Go back to the ETX Home Page.