[ceefna]

Hi Guys

I am such an idiot, I was testing the logic supply voltage on the L293 between pin 20 and ground (pin 16 on the schematic)and I had my multimeter plugged into amps and shorted something, duh.

The board still works, but there is a 10 second ish delay before it starts to track. Once tracking if I slew the motor to the right at full speed it takes about 10 seconds to start tracking again but if I choose a slower speed(you get 1 to 9) say 8 or below then it will track immediately.
Its like it is having to re-charge/re-buffer before it starts to track again.

Now if I slew the motor to the left then it will start tracking again almost immediately. so its a problem moving right.
I have found a schematic for the board but I am not 100% sure that it is correct as it is not a professional one, here is a link.The PIC's on the board are 16F876A and the L293 is 20 pin so the schematic is not perfect.
http://www.dd1us.de/Downloads/Schema...oard%200v2.pdf

I have checked the input to the L293 on pin 7 (which is the motor turn right pin) with a led and it flashes away ok when slewing the motor as soon as the button is released the motor stops and should start to track straight away but nothing happens. You can see the led is dimly lit for about 10 seconds then it jumps into life and the led starts to flash as it tracks. What could be causing the delay?
I hope you can understand this and have 5 mins to have a quick look at the schematic.
I think the L293 is ok as from my test it is not getting the signal to drive the motor from the 74HC240D for 10 seconds but what feeds the 240D. Could this be at fault? is it a lost cause trying to fix it? I don't know what to try next. I have found a new board on Celestrons web site for $89.00 but they do not sell outside the USA.
Thank you in advance for all your help and if you need any more info please ask.

Ceefna

[RomanLini]

Hi Ceefna, and sorry to hear you have troubles with that board.

Hmm, it's hard to offer any decent help based on a few sketchy symptoms. I'm not familiar with that product at all but I can see it has two(!) PIC microcontrollers. That means a weird symptom set can not easily be used to diagnose the fault.

In other words, if the micros go to do something and they sense there is a fault they will cause the symptom set, or delays or weird behaviour etc. If the circuit was just a couple of logic chips the symptoms would tell you a lot more.

In the situation that you don't have a lot of test gear or good knowlege of the board, it might be easier to do some chip replacement. Basically you could replace the L293 and 74HC240 (and maybe the other logic chips?), and if that fixes it, fine. If not it might be a damaged PIC?

Sorry I can't offer more help, but its not easy to diagnose microcontroller product faults based on symptoms, and a fault that occurred after a bad short makes it even harder as you can't predict a "standard" failure.

[HorridHenry]

I think the best option is to replace the chip,if you put your finger on the chip does it feel hot? bearing in mind not to touch the pins.

When I replace a SM chip I cut the legs off first and remove the remains of the legs with a sucker and soldering iron.this is all done under a good magnifying glass mind.

[HorridHenry]

Search: "L293"

oops! thats not a SM chip,that all it says on the chip L293?

Proper fella here.....
http://uk.rs-online.com/web/p/motion...ntrol/7140625/

[ceefna]

Hi Roman,
Thanks for the reply, I think the two micro controllers are there because it is an altitude and azimuth type mount so it needs to beable to mowe both the motors at the same time for tracking. In EQ mode it just moves the azimuth motor at the correct tracking speed.
I know this doesn't make diagnosis any easier but its probably only one if any of the pic's that maybe damaged.
I re-flashed the motor controll software onto the board and that went ok with no errors.
Is it possible it could be something like a capacitor? but i suppose it could be more than 1 part. That fact that it is still working(but delayed) makes me think it might be repairable.

Is there any way to test the signal coming into the octal buffer? Because the delay is there at the output of the buffer. I could see if the delay was at the feed going into the buffer that would count out the L293 and 240D as the problems?

Ceefna

[ceefna]

Hi, the chip is a L293DD, it's not getting hot at all. I have tested the signal into the L293DD when the delay is happening and it has no signal so I am thinking that the L293DD maybe ok? or could it be that the L293DD is burned and the signal needs to build before it flows? I am thinking like a plumber with a blocked pipe
I don't get why the fault only occurs when I move one particular motor to the right on it's fastest speed. Its like it's stealing all the power from the board and the components need to recharge before it starts up again

Ceefna

[HorridHenry]

The link I posted to RS componenets has a detailed pdf file of its characteristics.

[HorridHenry]

I've had a look at that chip and it has four inputs and four ouputs so the chances are one of the outputs is defunct or not working properly?

So with a meter check each input for voltage rise and see if one of them is not playing ball? then it could be that pins supplied input from where its coming from?

[ceefna]

I have checked input at pin 7(that is azimuth motor turn right input) on the 293 and when the fault is happening it has no input. Pin 7 is linked to pin 18 on the 74HC240D octal buffer and there is no output there during fault conditions.

How do I check if the 74HC240D is getting an input? which pin is the input for the output on pin 18?

Ceefna

[HorridHenry]

Cannot open the PDF file here,see if you get the same?
Buy Buffer and Line Driver Octal Buffer/Line Driver SN74HC240DW Texas Instruments SN74HC240DW online from RS for next day delivery.

[ceefna]

PDF working fine here, hers are 2 screen shots with logic info and pin numbers.
Am I right that I need to test the input at pin 2(for output pin18).

Can I test with an led or multimeter?

Ceefna

[doorknob]

I don't want to sound too much like a jerk here (!), but IIRC you bought an oscilloscope while you were building your drive circuit, and surely you will learn much more by probing the circuit with your oscilloscope than you will by using a multimeter or LED.

I understand that if you have limited experience using the scope, the thought of jumping into the deep end when you have a critical problem to solve may be unappealing, but without using the scope you may find yourself spinning your wheels unproductively for a long time.

Do you have a nearby friend who has experience using a scope? You can get help with that here, but writing down tips and instructions can require a long turnaround time compared with having someone in the same room.

[ceefna]

Hi Doorknob, how are you doing?
You are right I did buy a scope but I really have no knowledge on how to use it.

Is it safe to go probing around on the board? can I do more damage than I already have?

I am a mechanic remember, electronics are spells and magic

Ceefna

[doorknob]

I expect that it will be safe for you to probe around - the only real precaution to take is to make sure that you don't inadvertently short out two adjacent pins or circuit board traces with the probe, and that the ground clip of the probe is securely clamped onto a ground point and won't come loose and short out the circuit traces.

It is a good idea to set your probe to the 10X position (and remember that the voltages that are shown on the scope will be off by a factor of 10).

You can start out with the scope set to automatic triggering. You will have to adjust the horizontal timebase to an appropriate value based on the frequency of the signals that you are looking at.

BTW, I followed your build and thought that you did it very skillfully and professionally.

[ceefna]

Thanks for that, the other build was very successful.

I will have a probe about and see what I can find.

Ceefna

[Al_The_Man]

PDF working fine here, hers are 2 screen shots with logic info and pin numbers.
Am I right that I need to test the input at pin 2(for output pin18).

Can I test with an led or multimeter?

Ceefna

Note: pin 1 & 19 are enables for each block of buffers, so it either has to be a low on each enable or at least strobed low with a pulse.
If they are strobed it is hard to check with a meter, scope should show it.
Al.

[ceefna]

Thanks Al,
I will check both inputs are low and there is also an input at pin2.
If there is no input at pin 2 I will have to trace back to the next component.

Ceefna

[ceefna]

I don't think the link to the schematic was working.
Here it is again.

http://www.dd1us.de/Downloads/Schema...oard%200v2.pdf

Ceefna

[ceefna]

The 7805 voltage regulator feels quite warm?
I swapped it yesterday but it made no difference, could it be a component power supply problem rather than a faulty component?

ceefna

[Al_The_Man]

It would be easy to monitor both power supplies with a meter, the input to the 7805 and 5vdc output.
Especially when the error occurs.
Both sides appear pretty much identical, so it should be relatively easy to compare sides, assuming it is only one side that faults?
Al.