• Please review our updated Terms and Rules here

while we're at it - help repairing Model 3

R

rebeltaz

Experienced Member
Joined
Nov 9, 2008
Messages
195
Location
Shelby, AL
Ok... with getting my Model 4 running, I decided to pull the Model 3 out and see if I could fix it. I thought that the problem was in the monitor - that I could have fixed. But I'm not so sure that that is the case anymore.

When you turn the system on, the LED on the bottom floppy drive lights up and high voltage comes up on the monitor. I can see that the filament lights up and if I turn the controls up I get a bright raster with retrace lines. I just don't get anything on the screen. On the Model 4, it says something like Insert Boot Disk or something like that. I don't get any message on the Model 3. Can someone help me repair this thing please...

Thanks...
 
Your monitor

Your monitor

Sounds like the "fly-back" transformer may be shot. Unless you are a TV repair guy with a schematic it might be a tough road troubleshooting. I know you are looking for a quick fix. If you know some that works on TV's, you might want him/her to take a look at it. In any event, you have a high voltage problem. Could also be a filter cap.

Agent Orange
 
I doubt if it were the flyback. If that were bad there would be no high voltage and therefore no raster. Sounds more like either the boot ROM not doing its thing or poor connection of the video signal to the monitor.
 
rebeltaz said:
I don't get any message on the Model 3. Can someone help me repair this thing please...

Thanks...
Click to expand...

Try booting into Basic. Hold BREAK key down while booting.

Cass? should be displayed. Press Enter twice for Basic prompt.

Philip
 
Agent Orange said:
Sounds like the "fly-back" transformer may be shot. Unless you are a TV repair guy with a schematic it might be a tough road troubleshooting. I know you are looking for a quick fix. If you know some that works on TV's, you might want him/her to take a look at it. In any event, you have a high voltage problem. Could also be a filter cap.

Agent Orange
Click to expand...

Actually, I am a television service technician with twenty years of experience. Although I do not have a schematic (anyone? I would love to find a copy!), I know that the filter capacitor isn't the problem. That would generally cause either an severe hourglass shaped image or 60 cycle interference in the raster.

I'm not sure why you suggested the flyback or the high-voltage as a possible problem, unless it's because I made mention of the retrace lines when I turn the screen control up. Any CRT will do that when the screen control is turned up too high. But since cutting it back down to normal causes the lines to disappear, I don't think the problem is in the high-voltage circuit. BTW - my reason for turning the control up and intentionally driving the screen into retrace was to verify that the vertical circuits were functioning correctly.

Thanks for the attempted help, though....

Since only the bottom drive's LED lights up and not the top one, I'm leaning towards the system isn't booting up for some reason or another.
 
BillB3857 said:
I doubt if it were the flyback. If that were bad there would be no high voltage and therefore no raster. Sounds more like either the boot ROM not doing its thing or poor connection of the video signal to the monitor.
Click to expand...

pavery said:
Try booting into Basic. Hold BREAK key down while booting.

Cass? should be displayed. Press Enter twice for Basic prompt.

Philip
Click to expand...

I will try the booting into BASIC trick, but I'm leaning towards the boot ROM. Would that require a chip replacement or circuit troubleshooting?

Thanks...
 
If the system isn't booting at all you'll want a logic probe and multimeter to investigate.

First step: Check your voltages coming out of the power supply. If they are incorrect, fix and see if it boots.

Next step: Check pin 6 of the Z80 CPU. You should have a steady clock pulse. If not, check the clock generation circuit.

Third: Check pin 26(? Going by memory) of the Z80 CPU as you power up the system. The pin should be low for about 1/2 second then flip to high. If the reset line isn't being held low long enough then the CPU won't initialize properly. Fix the reset circuit if this is the case.

Fourth: Check the address and data lines. If they are dead, replace the CPU. If they are pulsing at a steady beat (different rates for different lines) that gets slower or faster as you go up the address lines then replace the CPU. Check the interrupt, read, and write lines. If they are dead, replace the CPU.

Fifth: Pull the ROMs and read them in an EPROM programmer. Check their checksums to see if they are good. If not, replace them.

Sixth: Pull the low bank of RAM and swap it with another bank. If you have bad RAM at the lowest bank of memory then the system won't boot.

Beyond that, it's a matter of checking the enable lines on the memory (ROM & RAM) to see if they are being toggled. If not then you have an address decoding problem.

Raymond
 
As you already know, it's not the monitor. By turning up the contrast and brightness to full, you see a pattern on the screen that fills the screen, and with re-trace lines (diagonal lines). That is normal for these machines and indicates that the basic monitor (TV) circuit is working. But it could be the monitor's video input circuit, or maybe the video driver on the logic board, right? Or, if the video circuit wasn't providing video, HSYNC and VSYNC, would you get the retrace lines?

A bad ROM isn't the first thing I would look at either. If you had a spare ROM, swapping it would be a quick test. But, lacking such a spare, there are better things to test first.

Based on your description, the CPU is working, at least initially. Since the floppy disk light comes on, it looks to me like the CPU is searching for a floppy. That light doesn't come on unless the CPU asks it to. The disk lights may flash on power up, but then a moment later the CPU looks for a floppy, turning on the LED for a few seconds. So in the basic sense, the logic board is working.


PM me and I can send you the service manual for the Model III. The manual has schematics and a pretty complete theory of operation.


In the mean time some basic tests to try. Raymond's tests above are best if you just want to probe the logic board, or try the following.

Did you get a CASS? prompt like described before? I expect not, but, if you did, then you know something is wrong with the disk sub-system causing the computer to lock up.

Do you have a working floppy disk to boot from? If so, put it in, and see if the floppy drive does more than just turn on it's LED and then go off. You should hear a few head movements. Also, I think TRSDOS 1.3 (maybe others) does a quick scan of the other floppy drive, causing the LED on the top floppy to come on.

If you don't have a working floppy disk, maybe you can do something from BASIC (boot into CASSETTE BASIC) that will show signs of life. Try making the cassette on/off lead transition with a CLOAD command.

To get into cassette basic, hold the BREAK key while pressing reset. Press Enter twice (well, 3 or 4 times just to make sure), and type CLOAD <ENTER>. If you have a cassette drive hooked up, with the play button pressed, CLOAD should cause the cassette drive to start playing. If you don't have a cassette, then probe the cassette REMOTE line from the 5 pin DIN connector to see if the line transitions. Info on the cassette cable can be found here:

http://www.8bit-micro.com/data-cable.htm

Alternatively, if you have a printer (that you know works), hook it up, do the same above to get into BASIC, and type LPRINT "Hello" <ENTER>. Hello should print on the printer.

I can't think of other BASIC commands that would show sign of life, so the above may be your best options.

Being a TV guy, you might want to check the VIDEO, VSYNC, and HSYNC. They are on connector J5. Video is pin 2 on J5 (or pin 13 of U52, the video generator), HSYNC is pin 6 on J5 (or pin 13 of U23), and VSYNC is pin 4 no J5 (or pin 12 of U23). U23 is a 74LS221, with 8uS given as the HSYNC value, and 693uS given for the VSYNC value (based on the RC circuits on the 74LS221 I assume).

Beyond that, it boils down to the kind of tests Raymond suggests above. Raymond's tests might be easier than my suggestions being that the machine is already disassembled I assume.

I hope this helps.

Mike
 
Wow! Now that's a troubleshooting guide! :D

First, in answer to your question, whether the video circuits are working or not, you'd still be able to get the retrace lines.

It never occurred to me that the floppy drive wouldn't come up unless the CPU asked for it. I don't have a spare ROM, but from what you've said, I don't think that is worth the trouble.

I'll try your suggestions tomorrow. I really appreciate everyone's help.
 
Any success with this repair yet? Just curious.

Someone pointed out the other day on another TRS-80 forum that if you type CLOAD, you might be able to hear the cassette relay click (to enable the remote line). That's another way to look for signs of life without actually needing a multi-meter as I suggested earlier.

Over the weekend, I fixed a Model I that had screwed up video. It was an interesting task. I hadn't realized that half the circuit board of a Model I computer is just the video circuitry alone, all digital with a very simple 2 transistor analog output to create the NTSC video signal. Anyway, it was actually very easy to trace back and find the bad IC once I knew what to look for. The vertical sync was getting lost. I replaced the IC (a 74C04) and the video was working great. It got me thinking again about your machine and wondering if you had any success.
 
I wondered about that. The IC I removed was not the original IC. This fix has been performed before on that machine. Any idea why that particular part fails a lot? Other than being CMOS instead of TTL (more susceptible to static?), I couldn't see any reason for it to fail more than anything else.

Yes, the 74C04 is a 4069 - a CMOS hex inverter, and the National part I bought was labeled as both. Thankfully they were still available at a local electronics shop, but DigiKey has them as well. I assumed that 74HC parts would have replaced all the 74C parts long ago.
 
Lohmeyer said:
Any success with this repair yet? Just curious.

Someone pointed out the other day on another TRS-80 forum that if you type CLOAD, you might be able to hear the cassette relay click (to enable the remote line). That's another way to look for signs of life without actually needing a multi-meter as I suggested earlier.
Click to expand...


To be honest with you - I haven't had time for my own endeavors lately - too much to do in the shop. I hope to get to it soon, though. Thanks for the thoughts...
 
Lohmeyer said:
I wondered about that. The IC I removed was not the original IC. This fix has been performed before on that machine. Any idea why that particular part fails a lot? Other than being CMOS instead of TTL (more susceptible to static?), I couldn't see any reason for it to fail more than anything else.

Yes, the 74C04 is a 4069 - a CMOS hex inverter, and the National part I bought was labeled as both. Thankfully they were still available at a local electronics shop, but DigiKey has them as well. I assumed that 74HC parts would have replaced all the 74C parts long ago.
Click to expand...

My guess would be static, probably caused by people touching the expansion port in the same manner that the 6521? chip in a C64 gets blown up sometimes when you touch the port it's attached to.

The horizontal driver chip (74C00) goes a lot too as well as hot RAM chips. All of which are static sensitive.

The newer Model Is Rev. G boards came with either 74Cxx or 40xx chips in them (I think the 74C00 is a 4000, but, don't qoute me on it, it's been a while)
 
That makes sense. I so often take it for granted that modern electronics is a lot stronger against static these days, _and_ testing is much more stringent to make ESD hardened products. I also see the Model III video circuit is no longer using CMOS gates, but instead uses a pair of 74LS221 IC's to generate and drive the horizontal and vertical sync. I'm always paranoid handling MOS devices like MOSFETs, 74Cxx, 4000 series, and RAM, but less so for BJT stuff like the 74xxx series stuff.

Also, whereas the Model I mixes the Video, Hsync, and Vsync together to make a signal that works directly on any composite monitor,the CRT controller board of a MIII takes Video, HSync, and VSync unmixed.

Sorry, kind of hijacked this thread. When the TRS-80 came out, it fascinated me how the video was created because it was all a mystery. Up until now I never spent the time to see just how ridiculously simple it is (and I'm an EE by trade). It's just a bunch of shift registers, and clock dividers, but I always thought it was some complex analog circuit.
 
I've always had all that nice anti-static stuff at work, but not at home. I just use other precautions which are effective enough. But that doesn't help when buying CMOS IC's at the local surplus store, and they don't have a complete anti-static environment. They just pull the part out of a bin and "try" to be careful.
 
You must log in or register to reply here.
Back
Top