TRS-80 Model 1 ... DOA ?

[pkasson]

Fellow 80-ers,

Just got a PS for the 80 as I thought the old one was dead. Its connected to original monitor which works, but when 80 is turned on, screen starts flipping and snow ... and no power light on keyboard.

How can I tell if the MB is bad - didn't see any burnt items and all the capacitors look good.

Thanks !

Pete

 

[ldkraemer]

pkasson,
First thing is to VERIFY the +5 VDC Power supply to 5.00 VDC, then the +12 Power Supply at 12.0 VDC.
(You can adjust the +5 VDC supply and the +12 VDC supply. But, don't try to adjust unless you know
they are not in current limit. - REF Hardware Manual). Then verify the -5 VDC Power Supply.
If those are correct, touch each RAM IC and see if any are RED HOT. If so, remove those and replace
them. Then check them again. Those RAM IC's are bad about shorting to the -5 VDC supply.

I have included this information in my email. Have a look at:
http://ts-inc.dyndns.org/Diagnose Repair TRS80 TRS-80 Model 1.html

It states to "REMOVE SHUNT Z3", but make sure you know how it is oriented before doing so.

The address lines are outputs from the Z-80 used to specify memory locations. If any address line is not active (stuck on one state), there will be 'garbage' (random characters)
on the screen at boot. To check the address lines, remove the DIP Shunt at position Z3 and power-up the system. If the screen fills with a pattern of “@9@9@9@9...”, then all the
address lines A0 – A9 are good. The remaining address lines A10 – A15 can be checked for activity with a scope.

This will give you some clue as to what is going on. It's likely a Power supply that is in current limit and folding back it's output.

Larry

 

[ldkraemer]

You may want to remove all the RAM IC first and VERIFY the Power Supplies (+5 & +12). If they are not at
+5.00 VDC and +12.0 VDC, then adjust the +5 first, then the +12. Set +5 to +5.00 first, then the +12 to +12.0.
Then verify that the -5 is close to -5. (not adjustable).

Larry

 

[Chuck(G)]

If it's of any encouragement, something as old and simple as a Model I is almost always repairable, save for catastrophic damage. I wish I could say the same about some of the modern gizmos that have gone south on me.

 

[Jay Newirth]

If you think you don't need the ps you purchased from me - send it back - will refund you
jay

 

[Dwight Elvey]

Hi Pete
If you will recall, I said that the power supply module ( a transformer, fuse and rectifier ) may not be the issue. You may have blown the power brick's fuse with a failed part in the computer. As a beginner, you may determine to use the shortcuts suggested by others on this forum. Larry's post is a possibility as it is relatively simple. It is not good trouble shooting but it may get by. Most RAMs do not fail short. It does sound like the symptoms indicate a short someplace. It won't show anything if you've already blown the power brick.
First you need to determine if you've blown the new supply brick that Jay sent you. I don't think it is fair to him to return him a bad blown brick that was damaged by you.
You can also use the other shotgun methods of just going through and replacing all the capacitors on the motherboard. This again is poor trouble shooting You may damage the mother board if your desoldering skills are not up to par. You may actually fix the unit. The problem may not be the capacitors ( although highly likely that one has failed ).
As Chuck says, these can be repaired. There are few parts on these that can't in some way be dealt with.
Trouble shooting takes some skill and some tools. A minimum is a good meter. For trouble shooting I prefer an analog meter with a real needle. Still, one can use a digital meter and if it is the only one you are going to buy, I recommend one. Even the cheap $10 Harbor Freight meters are quite useful. I have one myself. I don't worry too much about it. If I accidentally run over it in the driveway, the loss is minor.
We can help you some remotely from the this message board. It does require you to learn some as we go and when we make request for information about an indication, you need to be concise with the answers. It can be frustrating for both sides but a few of us are used to it. We will try to be concise with our request for information.
Dwight

 

[troj]

Reading this with interest, as I have a Model 1 which was working, but since does not. I'm fairly certain it has a short somewhere, I think on the high voltage side as that's the fuse that blew when I tried to use the modern power supply I got from Ian.

Suggestions on where/how to start diagnosing the issue? I do have a multimeter and I'm reasonably competent with it.

-Kevin

 

[Dwight Elvey]

Finding a short is actually relatively easy. Locating the actual part is a little tougher. If you have a bench supply there are some tricks on can use.
There are two main rails in the TRS80 from my view of the power circuit. There is the 12V and the 5V. With the ohm meter, you should be able to find the rail with the short. If you locate the path the power distributes on, I have a method that I've used that doesn't communicate well in words.
One method is to use the bench supply. These typically have a current limit or voltage set. I recommend placing the voltage at about 1 volt. Place the leads on the power rail that has the short. Slowly increase the current limit and feel around for something getting hot.
You don't want to hit it with the maximum current right off as it may open. Locating a damaged open part is tougher than locating a shorted part.
The key is having a bench supply with an adjustable current limit.
The 12V and 5V are the most likely locations of the shorted parts. Other places can cause input over current. The input bridge rectifier can have a shorted diode. If these are attached to the transformer, they are hard to test with a ohm meter, because the secondary of the transformer looks like a short. Isolating which diode failed is difficult without removing one lead. ( I use a scope and a variac but that is best done by showing on a bench ).
Another place that can fail is that the pass transistor in the regulator has gone short. This is bad as it will usually damage one of the ICs as well. You can test the pass transistor with an ohm meter as well but usually requires being removed from the circuit. ( again, I use my variac and scope ). A transistor looks like two diodes with an end of each tied together, with an ohm meter. In one direction there should be close to infinite resistance and in the other some non-short resistance. How much resistance depends on the ohm meter used. A typical diode will have anywhere from 0.2 to 0.65 volts forward drop, depending on the diode type and the current flow. Some meters have a diode scale. These usually read in volts.
Dwight

 

[troj]

I don't have a bench supply. I'm finding DC ones on eBay for reasonable prices, but nothing AC.

-Kevin

 

[Dwight Elvey]

I don't know of any AC ones. Bench supplies are DC output as a general rule. For AC, I mentioned a variac. This is a generic name I use for a adjustable auto transformer. The name variac belongs to a specific manufacture of auto transformers. I believe you'll find a bench DC supply most useful in debugging things. A 0-15V and 0 to 3 amps is a good size to look for.
Dwight

 

[shank]

an o'scope is your friend in situations like this...

 

[Dwight Elvey]

With a scope and being able to do differential inputs, you can easily check diodes rectifiers in action.
Yes a scope it a very powerful debug tool.
Dwight

 

[shank]

Dwight is right, but you can also check address and data lines for activity and proper TTL levels—can’t do this with a DVM or even an analog VOM as the signal transitions are too fast. With a dual channel scope, good probes, and such you can set one channel to watch a clock and another to watch a signal such as RAS, CAS, or MUX and check the timing with the timing diagrams in the various tech ref manuals... and many other timing dependent signals that can help you find problems... the only way to see what’s going on is to use a scope. Logic analyzers work great, but good scopes can be found used for not a lot of money and some of the newer USB based scopes and analyzers are not expensive now. On these old systems you don’t need a lot of bandwidth and this keeps costs down. Digilent makes some good test equipment like this for reasonable prices. There are others, of course...

A scope with a bandwidth of 35 MHz is adequate for a Model I and probably through the Z80/68K based systems, 60 MHz is better. You want dual channel and to be able to trigger on channel A or B as well as to have a separate trigger channel (lets you trigger on the clock and watch 2 other clocks or signals on the display together so you can see the timing relationships). Be sure the unit lets you set the trigger slope like leading or trailing edge... Get good x10 probes and don’t cheap out on these—you need 3. X10 probes load the circuits less so show more accurate results—you can also calibrate them yourself with the scope’s “cal” signal on the front panel.

There are good scope tutorials on YouTube as well as in text form for free on the web. Techtronix used to have a very good one and I bet it’s still out there. Technical Support put out a good video training overview on test equipment and it includes scopes (link posted in another thread in the forum). If I could only have one bit of test equipment to check these systems with it would be a scope. All of this is why RS field service techs lugged them around to customer sites!

-kb

 

[Dwight Elvey]

The OP doesn't seem to have any voltage to light the LED when he turns on the computer. I don't think a scope has much advantage over a voltmeter to analyze what is most likely a shorted component.
Still, if someone said I had a choice to either keep my meter or my scope, I would keep the scope. Please stay with the symptoms.
Dwight