Sanyo M9998 Restoration.

[Superduper]

It is unclear what the SM means by D2, even sanyo doesn't mention an exponent type of suffix for that transistor and they are the manufacturer so I am pretty sure they mean 2SC930D. Perhaps they are enumerating the transistor of that series in the circuit if you have more than one 2SC930 in use. Tuner circuits are very finicky with transistors and I'm virtually certain that a 2SD930 will not work as a suitable replacement for a circuit where 2SD930D or higher is prescribed, otherwise they would have simply prescribed the non-suffixed version. I also mentioned previously that if you are buying your parts and they come from China, they will most likely be re(labeled) generics and not genuine transistors. This will be especially true if your replacement and originals do not look exactly the same with exactly the same fonts, etc. There is a reason why there's literally a million different types of transistors, the characteristics are carefully chosen for the application where they will serve. I'm not saying this is your problem for sure, but I've tried using the general non-suffixed version of transistors before in tuners and have not been successful in getting them to work in that circuit. You can try to find a transistor with similar characteristics and see if those will work but generally speaking, this is going to be a trial and error deal, and many replacements with similar characteristics might have different lead positioning so you may have to twist the leads to get them to work. As for whether "any" 2SD930D work better than a 2SC930, the specs state that the "D" version has 50% higher amplification. If you circuit needs it, then yes, it makes a difference, but again, it needs to be a real and not counterfeit one. I have purchased many transistors off eBay over the years and literally NONE of the ones I received tests with the correct characteristics in hfe, capacitance, etc. as compared to the datasheet. Even more disturbing is that if I purchase a batch (of 50, for example), tested individually, the results are all over the place, which in my opinion, renders them useless for circuits where specifications matter. They would only be useful for switches or other non-critical applications.

2SC930 Conv in the parts list is merely clarifying that the transistor is functioning as a mixer-converter. Ignore that portion.

 

[SGrant]

I think I mistakenly typed 2SD930D instead of 2SC930D.

Here's a photo of the original Q104 transistor. How do those characters translate into what number I should look for?

 

[Superduper]

That is a 2SC930D, definitely not the derated 2SC930 or 2SC930C.

 

[SGrant]

This is the closest I could find. Also the most expensive. Should I risk ordering one?

2SC930 Original Sanyo Small Signal Bipolar Transistor | eBay

Let us fix the problem for you. We will do our best to get back to you as soon as possible. - Parts labeled as "pulls" or "pulled" have been previously installed and are made by the original manufacturer.

www.ebay.com

 

[Superduper]

I suspect that the Dalbani offering is a real OEM one. However, although the photo shows a 2SC930D, the description does not list it as such. So you should confirm before you buy. Otherwise, they will say it was accurately described.

Should you get it? I'm not going to read through 5 pages of posts on this thread, but I'm unclear why you changed the transistor in the first place? Did it test bad or did you just suspect it is bad. Even if it's bad, it's also possible for there to be other issues too. Essentially, it's important to recognize that a replacement part will only fix a problem if that part is the cause of your problems. Before you plunk down $11 plus shipping for a single transistor, I recommend you first test the old part and your replacement to see how they compare. Regardless, if that part is bad or not testing to spec, then you'd need to replace it before moving forward. Generally speaking, diagnostics often require you to replace parts one at a time during dynamic signal analysis.

Cheap component testers like these are all over eBay or Amazon or Ali. I'm talking about the ones similar to these: https://www.ebay.com/itm/395422128834?

I'm not recommending this one, or any one in particular, just to show you an example that for under $20, you can buy a tester like this that will give you instant out of circuit tests on many components. The important thing is that you can tell at a glance whether the component is a transistor or other component. It will tell you whether the transistor is a PNP or NPN. It will tell you the pinout based on how it was inserted into the socket. It will also give some useful information such as hfe (beta) and vF (forward voltage). Although I wouldn't trust these measurements for critical applications, I have found that these are very useful for comparing transistors with each other. Compare the beta and vF with the datasheet to make sure that they are testing to spec. If hfe is supposed to be 200 but you are measuring 80, you know you have a potentially problematic replacement. Although greater detail such as capacitance of the transistor isn't noted, the other specs are good for ensuring that replacements at least tests to spec. I purchased a lot of 50 2SC2240 off ebay a long time ago and found that they didn't nearly measure what the datasheet specs them at. I tested them against each other too and found massive variations between the lot, with a few doa. This of course wouldn't be the case if they were authentic. Some slight variations are expected, but not by a lot and you shouldn't be able to find several, much less 1 doa transistor in a lot. If you look at the silkscreen layout of the eBay "cheap" transistors vs the original or OEM Dalbani ones, you'll see that they don't even match which is a clear sign they are just restamped generics or otherwise counterfeits. Forget the "genuine" or "original" claims, they lie all the time.

These testers are good for testing capacitors, resistors, diodes, transistors, SCRs, etc. They can even identify mystery components. If nothing else, it's a $20 tool that you can spend time playing with. The important thing is that I've found that the tester is pretty consistent, so it's good for comparing 2 parts to see if they test similar, good for identifying real vs counterfeit parts. You can even determine sometimes if a diode is a germanium or schottky diode. Just remember that if you are comparing 2 transistors, to handle them with tweezers or gloves. If you heat up the part with your fingers by different amounts, the results will vary too. Yes, finger heat will definitely alter a transistors electrical characteristics. In the end, you might find that the original part tests fine and within specs too. In that case, you can just put it back and save $11.

 

[SGrant]

I'm happy to locate any recap of this discussion so you don't have to waste your time reading through it.

As per post #52 the problem arose that the labels for the legs for Q104 on the circuit board do not correspond to the data sheets for the 2CS930. The leg I was touching could have been the base or the collector. Post #59 has more specific details. Post #66 has a photo of that part of the board and Q104.

Post #68 summarizes the situation:
"The AM band works ok with a vom set on resistance and the leads connecting preamp transistor Q104 - Collector to ground. I can change volume and stations using the radio's controls.

What may have led me seriously astray is thinking the vom was acting as an antenna. I stupidly overlooked that the vom has no way to select stations or change volume. And the vom was not connected to the base of Q104, which is connected to the AM antenna.

So. The radio's antenna must be connected correctly, S101 must be connecting the AM band properly and the main amp must be working properly. This is supported by the fact that bridging every associated portion of the circuits (in case there is a break) using jumpers did not fix the problem."

To me, that makes Q104 the prime suspect.

Post #76 has the schematics.

_____

Testing voltages on Q104 may be inconclusive because the values on the schematics may be with the AM or FM bands selected. The schematics don't detail which it is.

Here's my guesses, assuming the base is mislabelled on the board.
Guess #1: the voltages for Q104 on the schematics are for the FM bands selected.
Guess #2: with the AM band selected the emitter voltage on Q104 should be something like 6.5v and the base voltage should be 0.6 - 1.14.
(You went over this in Post #70.)

I'll go ahead and buy the tester you suggested. At least I can verify the two 2SC930's that I already have, with it. There will be a delay ordering it..

 

[SGrant]

I've ordered the M328 tester. The only local supplier is out of stock so delivery will take at least 3 weeks.

 

[Superduper]

While you are waiting, if you have the transistor out, I would suggest taking voltage readings again with the transistor out of circuit. You should ascertain which pins are which by tracing to the next component based on the schematic. Also make sure that the negative probe is at a good ground location or your readings won't reflect the true voltage. Also do not connect it to a ground on a different pcb.

 

[SGrant]

Please feel completely free to disengage if this matter is too frustrating to continue helping.

The original Q104 2SC930 transistor is out, so I think I've tested it. Because the result is not what I expected, I'll detail the test.

I confirmed the pinouts from the data sheet. That matches the characters on the tuner board. It confirms that I got normal AM reception when I checked for continuity between Q104 B and ground. The circuit diagram connections for Q104 are beyond my very limited expertise, but it seems to me the vom was providing enough current to Q104 to trigger it. So either Q104 is defective or whatever provides it with current is defective.

With the vom set on "V" with a straight line and 3 dots above the v, set to 2VKΩ, the red lead on Q104's Base lead and the black lead on the Emitter, I got a reading of .006 on one test and .004 on another. Reversing the leads got readings of 0.

Base to Collector got .000. Reversed got .001.

Repeating any of the tests just got zeros unless I let the transistor rest for a while.

 

[Superduper]

When I mentioned that with the transistor out, it would be a good time to retest, I was referring to testing the PCB as before, but with the transistor out of the circuit. Although this may seem weird, it could explain some reading anomalies. Also, you need to be specific on the circuit diagram EXACTLY where you placed the probes, the meter settings, etc. Sorry but I don't trust that you are doing things correctly or placing probes at the proper places so I would want to know the exact spot where you placed the probes.

As for the test you just did that you described on your last post.... I read it, and read it, and read it again carefully. Either I'm misunderstanding what you are doing, or you mispoke, or you have a total misunderstanding about how to use the meter, and a fundamental misunderstanding about basic electronics.

Because you either tested the transistor "in-circuit" or "out-of-circuit". Either way, it's totally wrong. If it was tested in-circuit, and using the V function, you are not testing for voltage at the B/E/C junctions, you are testing for voltage drop across the transistor. But more concerning is that it sounds like you are testing the transistor out of circuit using the V function of the meter, in which case, you would have better luck squeezing blood out of a rock. The transistor by itself will not behave like a battery no matter how many times you test it or switch probes between the leads. The one-thousands of a volt to six-thousands of a volt you are seeing is electrically insignificant because they are just stray voltages. In fact, I can hold my fluke meter in my hands and depending upon which fingers, or where on my body, the readings can be as high as 0.025 volts, and it would change every time I shift the probes or retest -- the readings you got are signficantly lower. No wonder you are confused. If you are attempting to test the transistor out of circuit with your meter, you should be using the diode function instead but honestly, it won't tell you nearly as much as the other tester you've ordered, and which I suggest you use as soon as you get it instead of trying to use the meter.

I think at this point, I won't be able to assist you any more. You need to have a better grasp of electronic principles first, and I'm not a good teacher for that neither do I have the patience to do so, you'll find much better guides and tutorials for that on youtube.

 

[SGrant]

Fair enough. You've been more than patient.

I'm not sufficiently conversant with the terminology or such electronics to know that when you said I should test with the transistor out of circuit, I mistakenly assumed you meant testing the transistor, not the circuit. It did seem odd to be measuring voltage that way, but what do I know?

I was hoping that if we resolved the problem with this M9998, it could serve as a guide for the seemingly many others with AM band failures on these things.

In the meantime the M9998 serves by playing my digitized cassettes, starting as mp3's on my phone, using an FM transmitter plugged into the phone to broadcast to the M9998 tuned to an FM frequency, and output to external speakers. I can remotely control equalization, volume and song selection on the phone. Satisfying combination of old and new tech.

After the tester arrives (in mid-July) I should be able to do a sensible test. At least it looks like a no-brainer that's suitable for my level of expertise. If the original transistor is defective I can think about getting a compatible replacement.