Considerations

Damaged UDN2993B driver chip

Parts and equipment required

Removal of the UDN2993B chip

UDN2993B chip positions

Ribbon cable soldered to base of circuit board

L298 to 74LS04 pin connections

LX50 to L298 pin connections

How the final assembly looks (incl. some other useful improvements)

Circuit and "in-situ" testing

How (I think) the control system works

Considerations

A more frequently occurring, and desperate, plea seen on various Meade LX50 owners' messageboards is -

"Help! I have blown the main chip on my LX50 circuit board!"

I fully understand this plea, I did exactly the same to my second-hand LX50 which I purchased in October 2004 and fried the chip within the first month. I realised very quickly that this was not going to be an easy problem to fix. To be fair, the sole UK Meade repair firm was very helpful. Their engineer said they had a few spare boards in stock and I should just package up my scope, send it down to them and they would repair it. With hope and enthusiasm I did just that, my hopes to be shattered a few days later when they said the boards they had didn't work and that they would have to send off to the USA to get the replacement parts. This could take several weeks/months. They sent the scope back and I was scrambling about for a solution for the next 2 months.

I found the Meade LX50 group and posted a message on it. I got several responses, all of them helpful, ranging from suggestions and contact details of stores and individuals who might stock replacement circuit boards, to a previous message thread which referred to a recently designed circuit modification for replacing the driver chip.

Thanks to Beevo for referring me to the message thread and especially to Dirk van den Herik of the Netherlands for the modification details.

Please note that all credit for the original design and work goes to Dirk.

The only credit I will take is that I have applied Dirk's modification to my updated LX50 circuit board, and photographed and documented the work - not original at all.

The fix is extremely easy to carry out, the parts are readily available from any hobbyist electronics stockist, and, the whole job can be done for under £10 / 15 Euros / $18 (as at February 2005).

Just one disclaimer before I describe what I did below -

I cannot take any responsibility for any modifications you make following the hints, suggestions, and details on this website. My understanding of electronics is very limited and my interpretation of how individual components and the resultant systems work is solely my own and not that of a qualified and experienced electronics engineer. Any damage, or voidance of warranty, resulting from carrying out any of the operations described below is entirely your own responsibility.

*Dirk suggests that this modification can also be applied to the LX200 line of telescopes as they also have the UDN chip.

Damaged UDN2993 driver chip

Below is an image of the damaged UDN2993B driver chip on the LX50 circuit board. When I sent Beevo this image his e-mail back to me was "Ouch!" - an understatement I think! When I actually short-circuited the board (I carelessly shorted the battery pack while I still had an external 12V supply connected to the front of the board) I heard a distinct "Phwop!" sound. This must have been the top of the chip "exploding" as you can see by the arrow pointing to the 2mm x 6mm hole on the top of it.

Parts and equipment required

 

L298N dual full-bridge driver

 

74LS04 Hex-inverter

14-pin DIL (dual in-line )socket

(for mounting the 74LS04)

Circuit/matrix board (plain)

Ribbon cable

Fine tipped soldering iron

Solder

Wire snippers

Hot glue & gun (not essential, but advised)

 

Circuit tester (DC voltage)

 

* click on the links for the detailed technical specifications *

Removal of the UDN2993B chip

I removed the chip by cutting its 8 pins at the top of the board using wire snippers and then lifted the chip up and waggled it about until the pins on the other side broke off. This left soldered pin stubs on the bottom side of the board to which I would later solder the ribbon cable terminals.

UDN2993B chip pin positions

 

The image below shows the old UDN2993B chip positions (as viewed from the top - component-side - of the board). You need to know these to solder the ribbon cable terminals correctly to the L298 component later.

Ribbon cable soldered to base of circuit board

 

I chose to solder the red lead to pin 1 and I coloured the tips of the other leads at both ends, using fibre pens, to aid identification later. The image below shows the ribbon cable soldered to the old UDN2993B pins on the under side of the board.

This one shows the view from the top side of the board.

L298 to 74LS04 pin connections

 

I connected the L298 to the 14-pin IC socket on the matrix board using 2cm-3cm long jumper leads I made from a strand of the ribbon cable. Don't place the 74LS04 chip into the IC socket until all the electrical soldering and assembly is complete (i.e. just before the testing stage). The connection matrix is -

Note - take care when working from underneath the matrix board that you are connecting the correct pins.

The remaining eight pins are not connected to anything.

The pin positions of the two components are shown in the schematics below -

L298 (front view) .................................................74LS04 (top view)

note: odd-numbered L298 pin positions are to the front

LX50 to L298 pin connections

 

I connected the LX50 board (old UDN2993B) via the already attached ribbon cable to the L298 on the matrix board. You can solder from below and/or above the board (it gets a bit "tight" once a few terminals are connected). I only took one strand through to the top side of the board but, with hindsight, it might have been easier to do more - whatever suits yourself and your own ability really. The connection matrix is -

* Join all GND cables together

** These were swapped round from van der Herik's original modification. This now defaults the RA drive to a Northern hemisphere (E->W) rotation. To default to a Southern hemisphere (W->E) rotation simply swap these connections.

How the final assembly looks

 

After testing (see below) I sealed my "snakes wedding" wiring on the matrix board with hot glue to prevent shorting or any of the connections working loose.

You will notice from the last image above that I have inserted an in-line fuse holder which uses a standard 20mm 1A fuse. I also had to snip the battery connections where they were originally glued to the board and resolder them to the board. The cables had frayed through and could short (did this contribute to my original short-circuit and fried chip?!) - might be worth anyone with an LX50 checking this out as a matter of course. I hot-glued all my solder connections for the same reasons given above.

Here is another improvement provided by Geoffrey Way who successfully fixed his LX50 circuit board using the method described here.  Geoff added additional protection to prevent re-blowing the circuit.  Geoffrey's method is presented below with accompanying image of the modifications he made to the board.

"Obtain a diode suitable for reverse voltage protection, Schottky diodes are commonly used for this purpose.  Keep in mind you may pay a SMALL price in slightly more power consumption by adding this to your setup, but it's WELL worth it to avoid a more expensive and complicated repair.  Example: 1N4003 Micro 1-Amp Diode, less than $1.  Trim the excess off the leads to suit your installation. With a sharp  knife, you must sever the circuit at the spot circled in the diagram.  The diode will re-establish this connection when you install it as shown.  The diode connection at the top of the diagram is to one of the legs of a capacitor. If you can, push the capacitor's lead further through the board by touching your soldering iron to that spot to expose some of the lead on the underside, for an easier connection of the diode at that end.  Pay careful attention to the correct orientation of the diode, as shown in the diagram. solder as shown, and that's it!   Now you have some protection from reverse voltage."

Circuit and "in-situ" testing

 

Before fully re-assembling the board, motor drive connector, and front panel I tested the modified circuitry using the 0-10V DC setting on my very basic multi-meter.

I plugged the phone connector end of the handset cable into the socket on the circuit board, connected the 9V battery pack to its connector, and placed the black test probe on the -ve battery terminal, then switched on. The LED flashed a bit then stayed on, the handset LEDs scrolled round a couple of times and then stayed lit on the 32x speed indicator - looking good so far!

I then used the red test probe to touch the pins indicated in the LX50 to L298 pin connections table above, while pressing/not pressing the various handset buttons. I noted the readings and they seemed to make sense. I was a bit concerned about the pulsing of pin 2 on the LX50 board when nothing was being pressed on the handset but I then realised that this was the average voltage pulsed signal which would be sent to the RA drive all the time and keep it tracking at the Earth's rotation speed.

I switched everything off and loosely tied the new L298 circuit board to the ribbon cable with insulating tape (just to neaten it up and make it easier to insert into the control panel housing later). I then attached the motor drive connector to the 5-pin plug on the LX50 board. I switched on - the LED flashed and stayed on, the handset LEDs scrolled a couple of times and stayed lit at the 32x speed indicator, a short "whirring" surge could be heard from the RA motor, then - a reassuring constant whirring and clicking from the RA motor - it was tracking! I pressed the E and W, N and S handset buttons, changed the speed and repeated pressing the direction buttons - total success!! And it has remained that way!!

!! Thank you very much Dirk van der Herik !!

How (I think) the control system works

 

I had never heard of pulse width modulation before this problem with my LX50. When I short-circuited the board and had no joy with the repairs by the main UK agent I was desperate and thought I could simply use a potentiometer to control the speed of the RA drive motor. This worked a little bit but I couldn't get slow enough speeds - the voltage being delivered didn't have enough "oomph" to overcome the static momentum of the motor. I surfed the web and started to find various articles, adverts, and tutorials on speed control systems and circuitry for DC motors - mainly aimed at the hobbyist robot market. From all this data I think I now have, at least, a fundamental understanding of pulse width modulation (PWM) for DC motor speed control.

Here is one of the best descriptions I could find. "PWM is a common technique for speed control. A good analogy is bicycle riding. You pedal (exert energy) and then coast (relax) using your momentum to carry you forward. As you slow down (due to wind resistance, friction, road gradient, etc.) you pedal to speed up and then coast again. The duty cycle is the ratio of pedalling time to the total time (pedal + coast time). A 100% duty cycle means you are pedalling all the time and 50% only half the time.

PWM for motor speed control works in a very similar way. Instead of pedalling, your motor is given a fixed voltage value (say, +5V) and starts spinning. The voltage is then removed and the motor "coasts". By continuing this voltage on-off duty cycle the resultant motor speed is controlled."

In th case of the LX50, you can regard the original UDN2993B chip, and also the modified L298/74LS04 circuit, as a sort of "distribution board". The supply voltage comes in to the chip (+9V) and is reduced to +5V which is then distributed to the internal circuits of the chip itself, to the motor drive connections, and to the "brains" of the overall system (the other chips and circuitry on the board).

The original UDN2993B chip is a dual H-bridge motor driver, the two bridges working independantly of each other (look at the symmetrical arrangement of the pins and their functions on either side of the chip). There is an ENABLE input provided for each bridge which allows PWM via the other "brains" circuitry on the board (independant speed control of the RA and DEC motors). There are separate PHASE inputs for each bridge which determines the direction of the current flow, i.e you can change the direction of two motors independantly (RA and DEC drive direction control). This chip also had built in current-sensing and control circuitry which also prevented current flowing the wrong way (cross-conduction) when there were changes in direction of the motor (phase), generating current like a "dynamo", which would destroy the sensitive chip circuitry. In the case of the L298 chip, it doesn't have this built in current flow protection and you need to attach a Hex-inverter, the 74LS04, to it. Think of the 74LS04 as a sort of very fast operating set of diodes (input 1/output 1, input 2/output 2, etc.) which only allow current to flow in the one direction.

L298

74LS04

Jean-Luc Duhamel has drawn a detailed LX50 circuit schematic which helped me understand the relationship/arrangement of all the board components and helped give me the confidence to carry out the work that Dirk van der Herik described. Jean-Luc has kindly given me permission to reproduce his schematic here. He gave me a few words of advice however. He said that the there may a few errors in the diagram so don't absolutely rely on it (refer to the disclaimer above) but it does give a good general indication of the board circuitry.

jean-luc duhamel;s site can be found here