Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Jul Fri 10, 2020 8:08 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 38 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Transistor Amp Issues
PostPosted: Apr Fri 17, 2020 5:36 pm 
Member

Joined: Oct Thu 15, 2009 12:57 pm
Posts: 79
Location: Harrisburg, PA
Attachment:
2020-04-17_8-44-23.jpg
2020-04-17_8-44-23.jpg [ 220.89 KiB | Viewed 1317 times ]
I posted this in the Phonographs forum, but it was suggested I might get more traction in this forum. ***discussion not permitted*** boredom I guess, but I decided to get overly aggressive in replacing components on a stereo transistor amp that was working. All electrolytic caps had been replaced a few years ago, but I knew there were some resistors that were running high or over tolerance. The right channel had started to get a little weak & did not sound quite as good as the left. Since it’s a royal pain to disassemble and reassemble I decided to replace all resistors and thought, while I’m at it, might as well replace all the transistors & diodes, except for the outputs. The thought process was hopefully I won’t have to replace anything again. (Very expensive germaniums on the outputs; all test good with multimeter diode check.) The left channel is working fine; the right channel is blowing one of the replacement transistors. Although I checked all new resistors before soldering in, I did not double check the transistors before soldering in. I verified everything I did was OK and replaced the transistor since I was not sure if it was good before soldering in, only to have it blow again. I have been over everything multiple times, verified the transistors & diodes are in correctly, all resistors ins correct places, no solder bridges on the PCB, verified transistors with diode check on multimeter, compared resistance readings between right & left channels & can find nothing that I did was wrong. I am handicapped since I cannot troubleshoot under power and perform voltage checks. All power is supplied through one harness and a multipin connector that would be very difficult to remove from the cabinet & would require major jury rigging to utilize. Following are replacement transistors used with the suggested replacements from Sam’s Photofact (Right Channel no.’s only); used same on Left Channel): Q10, Q13 (Ge) – GE2/RCA SK3004 = NTE102; Q11 (Si) – GE10/RCA SK3020 = NTE 123; Q8 (Si) – GE10/RCA SK3030 = NTE 123AP. The NTE 123 & NTE 123A’s are close except for package, which I was matching the original transistor package. The transistor that keeps going bad is Q11.


Attachments:
GE RC 7831 BWD-A Sams 864_6n.pdf [275.2 KiB]
Downloaded 31 times
Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Fri 17, 2020 6:29 pm 
Member

Joined: Aug Sat 17, 2019 8:55 pm
Posts: 61
You will want to make yourself a Dim Bulb Tester (DBT) so you can make some measurements under controlled power. A DBT can be as simple as an extension cord with one conductor cut and a socket for an incandescent bulb added in series. You would probably want to start with a 40 W *incandescent* bulb.
Attachment:
20200417_122416.jpg
20200417_122416.jpg [ 2.59 MiB | Viewed 1309 times ]


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Sat 18, 2020 12:11 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 16624
Location: Albuquerque, NM 87123
Your mistake was to replace all components, good and bad. That is not repairing; it is shotgunning.

Your only real option is too put everything back where it was and do the repair properly.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Sat 18, 2020 1:49 am 
Member

Joined: Apr Thu 09, 2020 7:33 am
Posts: 109
Direct coupled transistor amps are a challenge to isolate. Disconnect the collector of q10 and temporary solder a 4.7k 1 watt to the base of q11 and ground. Plug the amp into the DBT a very necessary tool to have when troubleshooting a direct coupled amp and screw in a 40watt bulb then you can measure voltage. Measure q11 for near correct voltage if ok then the problem is q10 and back. The secret to troubleshooting a direct coupled amp is to break the bias feedback loop by disconnecting a stage and stimulating a fixed voltage bias with an external resistor then you can make some logical transistor sense of it all but use a DBT current limiting bulb to prevent current buildup. Let me know what the voltage on q11 is. Even with the DBT it is still possible to damage a new q11 because it's not a power transistor. Plug your DBT in a switched power strip so you can switch it off quickly if the new q11 gets hot with your finger on it. Assume it will get hot and switch it on then off quickly a little at a time. Good ol transistor amps. Be sure q12 is a pnp and wired correctly and that the emitter circuits of q12 are good to help safeguard the chance of q11 getting hot. Let me know.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Sat 18, 2020 3:28 am 
Member

Joined: Dec Mon 23, 2019 5:52 pm
Posts: 136
Back in the day these GE amps were real pains! They actually had an exchange program for some of them because of the issues repair shops were having with them. We found you could sometimes get away with using one, maybe two, SK or ECG subs in one of these, but never any more all at once, always had to get exact GE replacements.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Sun 19, 2020 12:22 pm 
Member

Joined: Nov Wed 30, 2016 7:35 pm
Posts: 7257
Location: Sunbury, Ohio 43074
Well, it's too late now, but shotgun replacing every component in something like this gives you nowhere to turn if there is a problem.

The dim bulb tester could help. If it's a linear power supply in the amp, a Variac could help as well.. just lower the line voltage and it may give you the opportunity to do some voltage tracing, or even use a sine wave input and see what's happening along the way. This may or may not work, but worth a shot.

As noted, DC coupled amps are always a challenge, since something wrong anywhere has an effect everywhere.

SK, ECG, NTE semiconductors all have the same problem ... they are designed to replace A RANGE of other semiconductors, so they seldom if ever match all the specs of the part they replace. They just meet or exceed them. If, for instance, one of your replacement transistors has 3x the gain of the one you replaced, that could be an issue. Etc Etc. Another caution for using these, is ... is the basing the same as the device it's replacing? ECB or BCE? .....

_________________
Preserving the hist. of electronics, one boat anchor at a time! :)
https://www.bbtvtestequipment.com


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Sun 19, 2020 8:03 pm 
Member

Joined: Oct Thu 15, 2009 12:57 pm
Posts: 79
Location: Harrisburg, PA
Thank you for all the valuable information. Regarding the diodes, X13 to X16 appeared to be germanium diodes, but the forward voltage for all of these measured in the 0.56 to 0.59V range. The voltage seemed to be somewhere in between what you would expect for either a germanium or silicon diode, but a little closer to silicon. And yes, maybe it was not the best judgement, but I did replace them with silicon diodes not realizing how sensitive this type amplifier could be affected by even small changes in active components. I have not been able to find any information regarding this, but does the forward voltage of a germanium diode increase over time? These diodes did not have suggested replacements in Sam's Photofact; only GE part numbers, and I cannot find any information based on the GE part numbers. Please try to remember that although I do have some understanding of electronics, I do not have the vast experience with repairing electronics that many of you have. I am trying to learn and this was my first project, which was probably not the best to start with. In retrospect I should have only replaced the out-of-tolerance resistors and been done with it.

Unfortunately, it's not like there are many experienced, reputable, good repair technicians for vintage equipment. I have heard many horror stories of folks who have had equipment "butchered" by less than scrupulous repair guys. In the grand scheme of things I have not sunk a fortune into this and it has been educational, if nothing else.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Sun 19, 2020 9:26 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 16624
Location: Albuquerque, NM 87123
45 Nut wrote:

Unfortunately, it's not like there are many experienced, reputable, good repair technicians for vintage equipment.


Wrong. They are right here.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 3:27 am 
Member

Joined: Apr Thu 09, 2020 7:33 am
Posts: 109
I don't notice any germanium parts from the voltages I see on the schematic. Most typical bias regulation diodes are 1N5343 silicone glass. Germanium diodes have a breakthrough voltage of 0.1 and silicones have it at about 0.56. I believe all semiconductor parts will be typical silicone in this amp. Germanium parts are not as temperature stable as silicone and the resistance goes down as the temperature goes up. But this amp uses silicones. Don't worry about measuring anything without power such as ohms diode scale because for now it will measure good without controlled power testing. You have already made those tests so now let's see if we can make controlled power on tests. Those diodes are there to prevent crossover distortion and they are carefully matched to the base bias voltage of the four last transistors but let's not worry about ECG replacements right now it's too late for that but I think I can bale you out even if it might be all out of matched parts hahahaha I like a challenge. I will try to explain this as best as I can starting from the output going backwards. Ok the output of this circuit is capacitive coupled to the speaker through C7. This capacitor blocks DC and passes only the AC output signal excuse me audio output signal not to be confused with AC power. Sometimes we techs assume things get understood correctly hahahaha. BTW audio is a form of ac. C7 isolates the dc operating voltage in the transistor circuit from the speaker. This amp takes a minus forty volts and divides it in half to output a maximum peak to peak voltage of twenty volts at it's best ability depending on where the volume control is put. That's the goal of the output transistors so let's see if we can get them to have correct dc divided voltage on them between collector/ emitter before we can move forward on this effort to isolate the problem we have to move backwards but what did you get on q11 after disconnecting q10 collector and attaching a 4.7k resistor to q11 base to ground. You will have to replace q11 again. You picked a deusy to learn on hahahaha.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 2:33 pm 
Member

Joined: Aug Sun 23, 2015 6:01 pm
Posts: 1099
Location: NJ 08090
Chris---Minus 10 points for improper use of the word 'silicone'. See me after class.... :D


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 3:55 pm 
Member

Joined: Jul Wed 22, 2009 3:07 pm
Posts: 839
From the schematics, from the voltages on the B-E junctions, it seems that all the transistors, except the first one, Q2 and Q9, are Ge transistors.
Small signal Ge transistors, like the one that has failed for you, Q4 and Q11, are generally max 30V devices, and here the rail voltage is 45V.
The C-E voltage on the transistor in that position could reach the rail voltage.
So it is possible that your replacement transistor just did not have high enough voltage rating!

Regards, Peter


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 6:14 pm 
Member

Joined: Aug Sat 17, 2019 8:55 pm
Posts: 61
Since you didn't post the original parts list or even the make/model of where this is from... What are the output devices? Do you still have the old parts?


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 6:41 pm 
Member

Joined: Oct Thu 15, 2009 12:57 pm
Posts: 79
Location: Harrisburg, PA
Attachment:
AMP Transistors.jpg
AMP Transistors.jpg [ 354.31 KiB | Viewed 1079 times ]
As I noted in my initial post, the output transistors (Q12 & Q14) were not replaced. They are still original. Also, the transistor that is failing on the right channel is Q11, which is a NPN, silicon transistor → NTE 123. Q8 & Q9 are also NPN silicon transistors → NTE 123AP. Q10 & Q13 are PNP germanium transistors → NTE 102. The equivalent transistors on the left channel were replaced with the same NTE replacement transistors noted.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 7:25 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 16624
Location: Albuquerque, NM 87123
Are the NPN transistors really silicon? Perhaps NTE has made a mistake on this. If you still have the old transistors and diodes I suggest putting them back in and starting again.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 8:09 pm 
Member

Joined: Apr Thu 09, 2020 7:33 am
Posts: 109
Oops my mistake I was focused on the bias diodes. You're right the pnp are germanium transistors and npn are silicone.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 8:55 pm 
Member

Joined: Oct Thu 15, 2009 12:57 pm
Posts: 79
Location: Harrisburg, PA
Attachment:
GE-10_Transistor.jpg
GE-10_Transistor.jpg [ 90.41 KiB | Viewed 1064 times ]
Suggested replacements were GE-10 / RCA SK-3020 - Above from the GE Entertainment Semiconductor Almanac 1973.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 9:04 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4401
Location: Boston, MA USA
The NPN transistors (SK3020) are really silicon. NPN germanium transistors were rare and not very good. So it wasn't uncommon to see a mix of PNP/Ge and NPN/Si like this, although it made for circuits that were difficult to bias and stabilize properly. This direct-coupled circuit was a valiant effort but it seems to be before its time -- proper transistors were not yet available to support it.

The OPs only hope is to re-install all the original semiconductors, especially the bias diodes, and start all over again.

-David


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Mon 20, 2020 9:07 pm 
Member

Joined: Apr Thu 09, 2020 7:33 am
Posts: 109
I just looked up the max voltage rating on the nte123 and sk3020 and it's not enough for this amp try ordering a few 2n3019 from mouser. orbanp has the right idea. Tbirdkid that was funny I agree I had a huge brain fart. I'm 56 and maybe been fooling with this stuff too long but dc coupled amps mix me up sometimes I almost have to have my hands on them to get them working hahahaha. Most germanium power amps are transformer coupled at least the ones I'm used to. This is the first time I've seen one that uses npn silicones oops did it again hahaha in conjunction with germanium. No wonder it's very difficult. I don't know if I can help.


Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Tue 21, 2020 12:27 am 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 578
Hi,

One issue you are faced with understanding and repairing an amplifier is the way the schematic has been drawn. On top of that, if you are dealing with a positive ground system it is less intuitive. It pays to look up the transistor types as some may be silicon, others Germanium. These have different base to emitter voltage drops.

Over the years I have developed two simple methods to deal with this. One is to simply flip the polarity of the circuit (works well for circuits with discrete parts , transistors, diodes, electrolytic caps etc that are polarity dependent) and then re-draw the circuit into a form where is easier to see what is going on. Typically this is used to flip a + ground circuit into its negative ground equivalent.

If you want to turn a circuit into its opposite polarity form, it is simple: reverse the direction of the diodes, including any zeners, change the PNP transistors to NPN and change the NPN's to PNP's and reverse the direction of the other polarity dependent part, the electrolytics and draw the circuit out. I have attached two diagrams showing the typical process.

It really makes no difference to checking the amp, the voltages are the same at circuit locations, except for your real amp you have the meter positive probe connected to ground because you are measuring the magnitude of negative voltages, but they are now positive voltages seen in the circuit drawn in its "complimentary form".

Looking at the re-drawn circuit attached, you can now see that it is an amplifier that is as common as butter on mashed potatoes.

The output stage "devices" configured from a Darlington transistor at the top which pulls the output voltage toward the power supply rail and a Sziklai (complimentary Darlington) the pulls the output toward ground.

The Sziklai acts exactly like a Darlington, except its input voltage drop is only that of one base-emitter junction, rather than two. So to get a silicon transistor Szlikai to start conducting only takes around 0.6V , vs around 1.2V for the Darlington.

The bias supply must be able to overcome a total of about 3 x 0.6V or 1.8V to get both of the output transistors a little into conduction (if silicon types) to avoid cross over distortion (if they were germaniums, about 3 x 0.3V or 0.9V) and then a little more to provide a bias current that will cause some voltage drop across the 5.6 Ohm emitter resistors.

This is why there are the 4 bias diodes in series to overcome the total of 3 Vbe drops plus a little more. Diodes are used (rather than resistors) because the have the same temperature coefficient as the base emitter junctions, about -2.1mV/degC, so as things heat up the bias current doesn't run away or increase.

That is for silicon transistors typically, but if many of the voltage measurements on your schematic are correct, the base to emitter voltages (I think some of them are approximate), it looks as though Q10, Q12,Q13 Q14 are germanium pnp.

Q9 would be a silicon, notice the higher Vbe. Q11 transistor, hard to say as both of the Base and emitter voltages are specified as 20V which is obviously not correct. Looking at the bias diodes, this data suggests they are germanium types. Obviously if you replace parts they should be of the same type (or close) If silicon transistors were substituted everywhere the bias diodes would also have to be changed to silicon types. But this would be a risky move as the HF response of the amp would likely need adjusting to make it stable, so stick to original or very close parts. Germanium power transistors have an intrinsically poor HF response compared to silicon types.

There is the usual arrangement of a "booststrap" capacitor C6 to help the upper Darlington on at high drive levels. There is the usual AC coupled negative feedback( passed by capacitor C34 and R34 and R33) to the amp's input, then Cf or C35 is used to lower the HF response of the amp. When there is feedback around high gain amplifiers, there is they chance they can go unstable, if the negative feedback becomes positive with phase shift through the amp at high frequencies. So the response is rolled off (this a subject of poles and zeros not worth going into here).

The DC feedback that stabilizes the DC conditions of the amplifier passes via R35 and R36 to the emitter of Q9 , and the AC component there is bypassed off by C4.

As noted, since these common garden amplifiers are all DC coupled, the DC conditions of one transistor affects all the others. The first initial check is to test all the transistor junctions (in circuit) for shorts, or being open. And of course the diodes and resistors & caps. If anything looks suspicious it leg/s need to be lifted for a better test.

The main thing in fixing circuits is understanding how they work, this is how re-drawing them in other forms helps the brain untangle the mess, so you are then able to interpret the test data you acquire and then postulate a theory on what might be wrong there and look in that area of the circuit.

(You will note some extra diodes not commonly seen in this sort of amp X-19 to X22 . The application of these diodes comes into play in overdrive conditions. By creating a small stand off voltage in the emitter circuits it helps the transistor's turn off well on alternate half cycles and reduces the power dissipation when the output stage is pushed into switch-mode conditions)

Shotgun replacements of parts, as other have noted, is a very bad idea, that can introduce more faults (for example if you accidentally install a transistor with the leads incorrectly and mix up the e-b-c connections) you can then be chasing your tail for hours.


Attachments:
ampj1.jpg
ampj1.jpg [ 354.87 KiB | Viewed 1052 times ]
amp2j.jpg
amp2j.jpg [ 257.65 KiB | Viewed 1052 times ]
Top
 Profile  
 
 Post subject: Re: Transistor Amp Issues
PostPosted: Apr Tue 21, 2020 2:34 am 
Member
User avatar

Joined: Sep Wed 10, 2014 7:13 pm
Posts: 1169
Location: Urbana, Illinois
In this type of amplifier circuit it is often feasible to replace hard-to-find factory original germanium transistors with silicon. This also applies to the strings of diodes which provide bias for the output stage transistors.

It’s likely that all of the original parts were germanium (both transistors and diodes).

You will want to end up with “all germanium” or “all silicon” semiconductors in the circuit. It is unlikely that any other parts values (resistors or capacitors) will need adjustment.

Although this schematic looks unusual, that is due to the way it is drawn on the page. The actual circuit design itself is relatively standard for medium-power audio amplifiers.

-EB

_________________
I’m learning to be a leading-edge designer of trailing-edge technology.


Top
 Profile  
 
Post New Topic Post Reply  [ 38 posts ]  Go to page 1, 2  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 7 guests



Search for:
Jump to:  


































-->


Privacy Policy :: Powered by phpBB