Repair Briefs from Other Sources


1.4) Right channel dead in Sansui 2000X receiver

(From John Heinrichs ( |

Sansui 2000X receiver
Purchased from a friend for $10 if I can fix it. According to previous owner, right channel completely dead and left channel distorted. Chassis loose in wooden cabinet (obviously an earlier aborted attempt at repair). No manuals or schematics. This is a superb receiver, 85 watts/channel of clean, warm power. High-end unit in the mid-70s. Not only that, I had previously picked up a matching Sansui RA 500 analog reverb unit for $6 at an auction and wanted to put the two together.
I started, simply enough, by checking the output fuses. Right channel fuse missing, left fuse looks OK. Aha! Easy enough... Moved the left fuse to the right and put a new fuse in the right. Replaced the fuse, thinking this was gonna be a good story. Now to plug it in and see what happens. Right channel dead, but left plays great. That stuff about distortion must have been a mistake.

Next suspect - the power output transistors. Sure enough, two of the four transistors have a direct short from the base to the emitter. Pulled 'em out and hooked 'em up to a Heathkit transistor checker (they are 2SC1111s, so are NPN silicon). Leakage current at 100%. Being short on cash right then, I went to my parts stock and found a set of four MJ423s taken, I believe, from a junked laser power supply. According to the Motorola guide, they are also NPN silicon and have a significantly higher breakdown voltage (325 V) than the originals. This amplifier is a pretty simple design and certainly too early for the fancy Darlington push-pull amps we see today. So I'm pretty sure I can drop the MJ423s in with no problem. When I power up the amp again, still no change. With a gasp, I shut it down and pull out the replacements and check them again in case the real problem had fried them. They're OK,

Now I look a bit further. Are the low-value (0.5 ohm) power resistors discolored? No, but I pull out the meter for an in-circuit test, which will work since 0.5 ohms is much less than the total resistance of the rest of the circuit. One of the four resistors is a dead short. Again, I go to the parts bins and find a great 0.5 ohm replacement (3% tolerance, no less) but the leads are too short. So I solder on some lead extensions and replace the bad one. Power up - now I'm getting a very weak

Tracing the wires backwards, there's a small circuit board (in a card socket, thankfully) that looks connected to the power transistors. So I pull the board out. Aha! A large black discoloration surrounds what's left of a 100 ohm resistor (the only way I can tell the value is by comparison with the duplicate circuit for the left channel). The resistor reads open, so I replace it, along with a nearby 0.1 microfarad capacitor that is a little discolored and has cracks in the outer insulation. Put

I start looking around and find a transistor that looks suspect. A cheap bent metal heat sink is glued to the transistor case with some brown gunk. Compared to the corresponding component on the left channel, the one on the right has an even darker color to the gunk, which looks crumbly. So I try the transistor checker (I have an in-circuit tester, which I recommend highly) and it shows a leakage current of about 30%. This one I manage to find in my guides and it shows a 2N2222A as a replacement. I've got plenty of these from earlier dissections, so no problem. After swapping this transistor out, the right channel comes up much better on headphones. Sounds mighty fine, in fact. So I hook it up to speakers. Now the right channel is highly distorted, which increases with volume. When I hook the right channel up to my scope, the waveform is distinctly clipped and only the negative signal excursions show up. Looks like the waveforms I get with

As I recall, I spent the next four hours or so checking every component I could on that little circuit board. I even went so far as to remove each of the remaining three or four active components and switch them between channels. No change - still distorts under a load. Then I started to do a detailed analysis of the circuit itself - actually developing a schematic - and discovered that the connections at the board edge were symmetric. Well, thank those engineers back in the 70s! I inserted the board into the connector reversed and again found no change in the distortion. Now I'm just about pulling my hair out. At

Now to look at every possible connection in the signal path. Speaker switch checks out fine. Only thing left is the output fuse. Fuse still looks good but, just to make sure, I replace it. Voila! No more distortion. Now it hits me that I switched the fuses before I turned the thing on at the beginning. There really was distortion on the left channel... I've never seen a fuse do this before - acting just like a diode, but only when a

All that remained was to clean up 20+ years of dust and spiderwebs on the inside, lube the pulleys for the tuning mechanism, and dig up some replacement machine screws for mounting the chassis to the cabinet. The cabinet looks like crap, but I'll deal with that later and do the reverb unit cabinet at the same time. Now I've got the receiver set up in my shop, and it really does look cool next to the reverb. The sound is excellent, and I have to confess I like the look of the old analog radio dial. Best of all, all the parts I needed were in

Postmortem/lessons learned: First, never change anything before the preliminary evaluation. I would have saved a lot of time if I hadn't switched the fuses around. Second, it would have also saved time to trace the entire circuit first and find all the problems (like the fried components on the small board) right away rather than the progressive look/replace/test process I

1.5) Magnavox 4head HiFi VCR

(From Willis Chung (

Magnavox 4head HiFi VCR comes in with chief complaint that it won't load and eject tapes consistently.
A quick look inside with the cover off confirms this. The tape inside will partially elevate, but will not eject. The machine will also not play the tape, even when it appears that it is all the way down.
With gentle finger assistance, the rental tape is raised and ejected while the VCR is trying to eject it. A scratch tape is used to confirm that the problem is repeatable. An examination of the mechanism in the under-carraige shows a single toothed drive belt. No slipping here.
Thinking that the tape elevator mechanism was sticking, I clean and lubricate the elevator assembly. Function does improve, but if a tape is loaded into the machine after a few minutes of inactivity, the machine fails to load it fully and fails to eject it. Once the tape is ejected with finger assistance, subsequent loading usually resulted in correct function, with the machine going into the half load position, and then

Thinking that there was sonething mechanical that was sticking, I poked and prodded at a lot of the mechanical parts, finding nothing wrong, nothing sticking. Thinking it was a mode switch problem, I took the mode switch out, disassembled it, and cleaned it thoroughly. Thinking the machine was not sensing the position of the elevator mechanism, I disassembled and cleaned the elevator position switches and verified their function. Naturally, none of these things changed the behavior of the

Stumped, I let the machine "season" on my bench for a few weeks. I fix

After posting the problem, I received several suggestions. #1 Check the elevated pinch roller mechanism for binding. I had cleaned and lubed this already, but I took a close look at it. No problems there. #2 Check the

Hooking up a scope to the outputs of the switch mode power supply shows massive ripple on the +15 volt line. Aha! Tracking the line back, I find a capacitor (470mfd, 16volts!). Soldering a 470mfd 35volt capacitor eliminates the ripple, and the machine loads and ejects like a champ. Examination of the other capacitors in the power supply reveals one more 16volt cap on a 15 volt power supply line. I replace this one with a 35volt one as well. Seems like the power supply was designed to fail

1.6) TTX 1450 SVGA Computer Monitor - Dead

(From Wilkin Claude (, Edited by Samuel M. Goldwasser)

TTX 1450 NI computer monitor
Gradual loss of brightness over several months followed by total failure.
Plug it in, dead, maybe some smoke...

A TTX 1450 NI monitor is brought to me for repair. Unit is from a family member who reports that monitor had been taking more and more time to show acceptable brightness when turned on. Monitor had been shown to a repair shop (person lives 200 miles from my home) that right away asked $50 for an estimate and $125 since the person there said (without opening the thing up) that it was the flyback or the picture tube that was defective. When I spoke to the family member and he told me of the story, I told him to bring me the monitor when he would come to visit me later that month if he didn't mind the wait. He decided to purchase another monitor and said if this one could be repaired for not too much money, he would use it to setup a computer just for his kids.

When I got the thing and plugged it, I saw smoke coming from some diodes and resistors on the board. One of these diodes was probably short, it actually glowed red! The thing would not show a picture, the power led in front would come on then turn off after about 2 minutes.

Visual inspection showed a swollen cap, a blown TL431 and a resistor so burnt that the color code could not be read. These looked like they were part of the switching power supply. The swollen cap showed a value of 7uf when tested, it should be 470uf. I have hope that the flyback or crt might not be the problem here after all...

Some components to be replaced could not be identified: that burnt resistor and two zener diodes. Time to get a schematic.

Luckly, TTX has an office right where I live. That schematic is available for $15. No, it's not a service manual, it's just a legal size copy of the schematic. I find this a little bit ridiculous but I have no choice, several postings on the internet for the schematic or just to I.D. the components brings no replies for the info. I get the schematic...

Finally, I see the component values. I probe around some more, I see that 2 zeners (30V and 100V) in series in the power supply are short. I replace: 470uf Cap, TL431 regulator, 2 zener diodes and the burnt resistor.

I turn unit on. No smoke, power led on but no picture.

I do some more probing around. I find that the horizontal output transistor is bad and there seems to be a 100 ohms resistor (that what the color code looked like...) that's open. I find a replacement to the transistor for $17 and change the 100 ohms 2W resistor. That resistor is in series with the primary of the flyback and the transistor. I suppose it when bad when the transistor shorted out.

I replace and turn the thing on. Hey! It works!...But wait, smoke again...I look around and see that the 100 ohm 2W resistor I put in is burning up...Time to look at the schematic.

My bad 100 ohms (so I taught) resistor is not on the schematic. I look there and I find a 1 amp. fuse. Exchanges through e-mail with other technicians brings speculation that 100 ohms is just not right. It's too high a value. I look at the resistor several times under different types of lighting and it always looks like a 100ohms...

I'm not going to blow another $17 transistor. Time to call TTX.I speak to somebody technical over there. He says I can put a 2.2 or 3.3 ohm resistor there. I guess it was 1 ohm or something like that. I still have the thing here and it doesn't look that discolored and I've been reading resistors for a long time and it does look like 100 ohms.

I put in a resistor (3 ohms) and everything is fine. I'll have a happy nephew.

Comments: It's hard to determine what really happened here. Perhaps the 470uf cap changing value brought this on. Having so many components fail is rare...I finally realised that the 1 ohm resistor is probably a resistor-fuse sometimes found in TV's. I have had to purchase these, but they were rare and cost $8. I guess that TTX guy I spoke too doesn't put one in too, he just uses

Go to [Next] segment. Go to [Previous] segment.

Go to [Table 'O Contents].

Please check attribution for Author. Processed by [Feedback Form] [mailto]. The most recent version is available on the WWW server [Copyright] [Disclaimer]