Model 4 turns on, then doesn't

For the first time in 5+ years I powered up my TRS80 Model IV. Given the time period since last use I was fully prepared to have nothing happen or, worse, some black smoke escaping. To my pleasant surprise, the drives started whirring, and up came the "Diskette?" prompt. I can also break into the Model III prompts and get to a basic prompt. Only problem was the vertical hold being out. No problem thinks I there's an adjuster for that on the bottom side (or is that just contrast I think now after the fact?). Completely failing to remember where it is I turn off and (very gently) turn the computer on its side. Thusly located I put the computer upright and power on again. And now it's not working Instead of spinning indefinitely the disks spin for about 5 seconds and stop. Nothing on the screen.

I've opened it up and done a quick inspection and there's nothing obviously wrong. The screen is getting power; turning up the brightness fully shows thin scan lines. The two fuses (that I'm aware of) are intact. No bulged/leaking capacitors anywhere.All the point fixes on the main MB appear connected (though I haven't done a serious check there).

So my two questions at this point are:

1. Can someone suggest what the next step is to track this down,
2. Is the vertical hold adjusted off the extenal adjustor or, (as I'm thinking now) off the four pots on the board driving the CRT?

TIA.

Check voltages first, IIRC they're +5V, +12V and -12V.

The CRT adjustment is done with the internal pots.

Turn to maximum brightness and contrast
If you insert the bootdisc and close the door is it booting ?
Is the drive spinning for a longer time now.

As jltursan has suggested, check Power Supply Voltages first. Also, if your Model 4 has the Tandy Power Supply, remove the Power Supply
(4 screws) and check that all Solder joints for the Power Pins are Good. Lots of times those pins will have a small ring around them making
a loose connection to the Power Supply Circuit Traces.

The Power Switch on the Model 4 breaks the LINE (HOT) and NEUTRAL (GND) from the AC Pigtail to the Power Supply. You can VERIFY
the ON/OFF Switch, making sure it is working correctly.

You can also easily depress the RESET Button, making sure it clicks and isn't being held down with dirt/dust, which would cause a continuous
RESET to the Z80 CPU.

Next step would be to check all the CONTROL Lines to the Z80 CPU (Active LOW).
WAIT
INT
NMI
RESET
BUSREQ

Also verify the CPU has:
CLK
+5 VDC
GND





Larry

OK, so I've done some testing, which should be taken in the context of my experience in this area (or lack thereof )

So the Z80 is definitely getting +5V on the correct pin.

I am confident the power switch and reset button work as expected.

I unscrewed the power supply and the soldering on the pins looks OK to me. However the results of testing the voltage on each pin may argue differently. So, values starting at the ground end:

* +12V
* Not Connected
* -12V
* +12V
* +12V
* +12V
* +5V
* +5V
* +5V
* -4.7mV
* -14.5mV
* -12.6V

So, the next step is to re-do the soldering on the two pins giving readings in the mV - correct?

OK, just to fill in what I did while that post was in the moderator queue:

Not sure if the problem was with those pins, or with my multimeter usage. Regardless, I put some extra solder on, and now the last 3 all read around the -12V level.

With further playing I remembered that you have to be holding down break when you reset :blush: , so I am breaking back into Mod III mode fine. Disk behaviour is still different to start, but more understandable.

So summarising:

* Can break into Model III ROM boot fine.
* Disks still spin for ~5 seconds then stop if left. If i stick a diskette in, but don't close the door, I get the indefinite spinning and the "diskette?" message coming up.
* Resetting with a disk in results in non-boot, but I have no faith that the disk being used is still readable. I have ordered a set of disks from ianm so that I am starting with known good disks. But the lack of anything sounding like head movement doesn't have me hopeful.
* The $*%&$*! vertical hold!!! I can adjust things so that the hold is stable (though things are _very_ touchy). But try to close the case again and something has shifted things slightly so the vertical hold is out again - lines cycle at with a 10-20s period. I'm hoping that it shouldn't really be that sensitive . . .

The Tandy Power Supplies were prone to have a little circle around the pins solder joint (Bad Joint), and that causes problems. They are also prone
to blowing the .01 & .1 Rifa Capacitors as they age. (Mouser Electronics stocks replacements)

Your start sequence of holding the Break Key DEPRESSED during Power Up is correct. I always do that, and then I tap the ENTER Key twice to get
the Prompt. At that point I always do a "?MEM" (without Quotes) and it displays the available memory (48032 if I've remembered correctly).
From there I boot the Floppy by using the RESET Key.

It sounds as if your Floppy Diskette is corrupt, or you could have a build up of Oxide (from previous use of old defective Floppy's) on your
Head(s) and that will prevent the system from booting correctly. You may want to clean the Head(s) and see if that helps. After this many
years it's also a good idea to remove the Floppy Drives, clean the Drive Head Guide Rails and Heads with clean Cotton swabs dipped in
Alcohol. I also use a couple drops of Dri-Slide on the Plastic Guide rails where the Plastic rubs on the Drive Guide Rails. GENTLY move
the Head Assembly full travel on the Guide Rails to spread the Lubricant. The Head assembly should easily move on the Guide Rails.
That makes them glide along the Rails and doesn't attract Dirt. Dri-Slide is available at most Motorcycle Shops.

Larry

I'll report back once I've tested the drives with a known working floppy. (I'll probably do the cleanup regardless, but after testing). I doubt there's Oxide buildup; they were reading fine when last used and, with nothing since, I doubt Oxide could have built up. dust, other stuff, possibly though.

Meanwhile, can anyone give me guidance on the vertical hold? Should it be that sensitive? There are 4 pots on the board next to the CRT. One pot shifts the whole screen left/right, one does vertical hold, and the other two expand/shrink in the 2 axes (I think). So there doesn't seem to be a separate coarse/fine adjuster for VH. If it is always that sensitive, how do you actually get it stable?? If not, what should I be looking at?

Thanks

It may not apply, but TRS-Ian posted about a sync problem in this posting:
http://www.vintage-computer.com/vcforum/showthread.php?36680-TRS-80-4P-Video-issues&highlight=sync

Give it a read and see if it applies in your case, if you have a non-gate array.

Larry

Thanks for that link. I have a non-gate array 4, and it doesn't have C210. I see contradicting info; Ian says in that post that it affects non-gate 4s but Tim Mann says "Early (non-gate-array) versions may not have this cap; it does not appear in the Model 4 Technical Reference Manual, 1983 edition." Based on it supposed to be near U5, there's no C210 there. And this is an early 4.

Looking at the CRT board from the other side, I was able to identify the pots properly. Use of "Vert Size" with "Vert Hold" does make things easier (i.e. get it close with the hold, and playing with the size - usually increasing size - stops it). I now have, or did a few minutes ago, the 16-line version stable, if right on the bottom of the screen. Argh, no it's moving down the screen again.

Other suggestions?

Meanwhile the floppys arrived from Ian M today, and Drive 0 is definitely not reading them. It's noise is also a touch unhappy at times. I'm also seeing varying behaviour. Sometimes the disks run "forever" and I get the "diskette??" prompt, sometimes they run only for 5 seconds.

Hoping for a next round of suggestions.

Peter

Well, here is a URL to more information on your sync problem.

https://groups.google.com/forum/#!topic/comp.sys.tandy/sw2bw6trKzA

Looks like booting into Model 3 Mode is best for the Initial adjustment of the CRT Board adjustments, and then booting into 80 Column
mode for adjusting C210 Model 4 (C231 on Model 4P) as referenced in the above URL.

In your case it's likely a problem with a component listed in the above URL for your Model 4. I'd use an O'Scope to check the 12.672 MHZ Clock
Signal first. A likely suspect component could be the C111, 10 ufd 10V Tant. Capacitor tied to the +5.0 VDC line.



Can you boot the Model 4 if you replace Drive 0 with Drive 1? The Drive select pins are removed in the cable so no
modifications on the Floppy Drive are necessary. Place Drive 0 to the side for this test, and only use one drive.

Larry

OK, after some minor distractions (holiday, Cyclone, week without power, and so on) I've gotten back to this. I also had some side conversations with Ian Mavric which helped. So, as things stand:

* I cleaned up the connector on the CRT board.
* I had another look at the joints TB1/TB2 connectors on the power supply, this time using a magnifier. Found some cracks I had missed before. The soldering job was, er, shall we say, not the best, but there is solder over the cracks.
* The floppy drives are non-functional. Thank you very much mice.

So, my display problems are definitely on the Model III side right now. After cleaning and more solder, it is significantly easier to lock in the vertical hold. However, if I reset, holding break, "Cass ?" reappears at different locations on the screen, i.e. I align things so all 16 lines fit, then on reset "Cass ?" may appear at the top, halfway down, right at the bottom, or off the screen completely. And then after 5 mins it will start drifting up the screen again (interestingly, always up).

So does this sound like I need to go over the TB1/2 joints again? (I'm going to get a shorter tip on my iron; I'm not happy about how the heat drops off before the end) Or is it something else?

Thanks again for feedback.

Update - not much improvement

Update - not much improvement

OK, so I got a working soldering station again and properly attacked the solder joints on TB1/TB2. I'm happy that all visible cracks have been melted and, where necessary, extra solder added.

The result of this is that getting the Model III display stable is much easier. However:

1. Given time (10+ minutes) it will start drifting again (though it can drift either direction now whereas before it only drifted up)
2. On reset+break the initial "Cass?" comes up in a random vertical location; it can be top, bottom, middle, off the screen completely.

Note that this is still just the Model III mode. Thanks to non-functional drives I haven't tested Model 4 mode yet. Also, after my latest round of re-soldering, the drives will spontaneously come on for 1-20 seconds

Given that the bulk of the computer is still functional, but there are other things wrong (e.g. the drives) I'm close to the point where I give up and pay someone with more experience to look at it and hopefully fix it without accidentally breaking something else. But I'd like to check a couple of things before then. Is it possible for power supply problems to cause the random vertical alignment of the screen on power up? Or is that something completely different? Do people think it's worth getting a known good power supply? Or is it into the "too many things wrong at once" category?

Thanks

pjhacnau said:

[*]Given time (10+ minutes) it will start drifting again (though it can drift either direction now whereas before it only drifted up)
[*]On reset+break the initial "Cass?" comes up in a random vertical location; it can be top, bottom, middle, off the screen completely.

Click to expand...

Hi,
Your display problems look similar to what would happen when some components are warming up. Try freezer spray / heat from hair dryer to find the area with the failing component (probably a capacitor).

Good luck,
John

Are you sure the drives are really non-functional?, just in case try to boot the machine without floppy controller. Disconnect it from the main board connector and check if it automatically boots to Model III mode.

Recently my M4 has failed and seems that the faulty bit is located in the controller (same symptoms as yours). Mine seems stable in M3 mode tho...

Your flaky behavior looks like a problem related to dry capacitors. Replacing them could be a good idea.

Just to finish the story off . . . I've conceded defeat, and by the time you read this the machine should be well on its way to TRS-ian to do a proper repair job on (which includes the drives, and wiring harness as well as the screen issue).

Ouch

Ouch

OK, so the story isn't quite finished . . . Ian is giving me a photographic log of the work, with commentary. And explains why the floppies might not be working The brown stuff (which Ian thought was probably tea/coffee) is produced by mice. And that's as much explanation as I want to give . . . the display problem has been isolated to the video board (not the power supply).

If anyone's interested in me continuing the story to the end (hopefully a fully working Model IV again) reply here and I'll post the condensed version here periodically. Otherwise I'll leave it with this.

I think you should carry on, it's an interesting story. I was going to chip in with my own M4 experience. I had the same rolling screen as you in Model 4 mode, intermittently. But I was able to adjust the horizontal hold to settle it down. Am now looking at the other posts to see which capacitors should be swapped (I have not done any work on it save fiddling with the h-hold and fitting a LoTech IDE adapter; might be time to check the power supply joints and caps). As it happens it is out of action right now because I scavenged a chip off it (74LS166) to help fix another machine. Replacement is on order!

Mice brown stuff?, bleurgh, I've always loved gore stories , keep on telling us about the repair Ian is doing.

jltursan said:

Mice brown stuff?

Click to expand...

I still have a mild case of PTSD from having to clean a mouse nest out of my TRS-80 Model I's expansion interface. Amazingly there was no serious damage from those little @#%$#s making a horrible mess on the non-component side of the circuit board, it might have been (effectively) terminal if Radio Shack had mounted it component side up.