ARCHIVED FORUM -- March 2012 to February 2022READ ONLY FORUM
This is the second Archived Forum which was active between 1st March 2012 and 23rd February 2022
Hello everyone!
I am currently restoring a Beomaster 4400 type 2419 (US). I have restored several European amps including a Beolab / Beomaster 5000 set in a recent past, but this one is the first of the late 70's.
Overall the project is progressing well and I am probably 50% of the whole project.
However, I’m diagnosing a rather odd problem and I would need guidance to clarify this mystery.
Here we go, so far, I changed all the electrolytics, some MKP that were seriously damaged and also changed the resistors (1% tolerance) around the quiescence trimmer to ensure its stability.
I started the amplifier, without any smoke… , to find an Overload fault. After a quick diagnosis, two TIP141 are damaged on the right channel. To continue my tests, I unplugged the power rail of this channel only to get rid of the fault and finally focus my tests on the left channel. I began to check the recommended voltages on the schematic and to note that all the values were way off besides seeing that the quiescence potentiometer always remains at 0mV despite having completely turned from left to right.
Instinctively, I reconnected the faulty right channel and checked the voltages of the left channel again without changing anything else and VOILÀ! All the voltage values were correct and the quiescence potentiometer started to respond correctly to rotations.
If I understand it correctly, when there is a fault, the relay cuts the -35V power supply rail on both output stages.
I’m under the impression that when the circuit is at fault, so without -35V on the left channel, the circuit makes its way to the ground and from there the voltages are correct, however, when the -35V rail is restored, the circuit goes crazy and the readings are all over the place.
Here is in detail parts changed and checked components:
Changed all the electrolytic (polarity followed from the old ones)
Changed some of the MKP
Resistors changed:
R160/260
R163/263
R165/265
R166/266
R169/269
R170/270
R171/271
R172/272
R177/277
R188/288
Transistors:
TR115=OK
TR116=OK
TR117=OK
OTR100=OK
TR119=OK
TR120=OK
TIP141/146= All good
(I didn’t check the right channel yet… other than the TIP141/146)
Anyone can give me direction where to look?
-------------------------
(Side subject)
0 0 1 504 2879 Compass-NAV inc. 23 6 3377 14.0 Normal 0 false false false EN-CA JA X-NONE I’m reading a lot of failed output stage transistors despite of the fairly well designed protection circuit. This is very curious, in my opinion. These are quite powerful chips but looking at modular heatsink; I’m really questioning the effectiveness of that design. The transistors are mounted on a thin 2mm back plate where four extruded heatsinks are mounted. The extruded heatsink contact surface is not perfectly flat and the surface contact area between the two pieces is very limited so the heat transfer from the back plate to the heatsink is very, very bad. From the test I made, the 2 little screws holding the extruded heatsink to the back plate are not enough to straighten the extrusion to increase the surface contact area. Additionally, the thermal probe, reading close the back plate, is not touching the back plate so the internal temperature must be very high before shutting down the output stage. All in all, IMO, the output transistors are running way beyond their recommend internal temperature and the cause of unexplained output stage transistor failure. Pictures to follow to discuss the subject!
Just a note:It's never a good idea to use metal oxide resistors in DC-coupled output stages like this.Just like C-type (high-HFE) transistors should be avoided too.There's a risk of self-oscillation.
If you fitted metal oxide resistors in the amplifier circuit where B&O didn't, I suggest you replace them back to standard 5% or 10% carbon resistors, and it'll be fine.
And there's no need for "increased stability" or sharp tolerance components anywhere in that circuit, - the trimmer adjusts and compensates for any off-tolerance components anyways and it's in itself a fairly high-tolerance component.
Output stage transistors rarely go bad without dragging more down with them.Check the driver, the emitter resistors etc. and don't forget the remaining output stage transistors as they will haveexperienced overloads and may easily have developed scars that will show sooner or later.
Martin
Jean
You don't mention any diodes in your list above, when I blasted an output channel on a BM4400 Martin recommended replacement of the following components........
I suggest:
Replace:0IC200, 0IC201, 0IC202, 0IC203TR220R277R276C238At least check (ideally replace):0TR200TR215TR216TR217TR219D204D205R271R272R278R279Martin
Dillen: Just a note:It's never a good idea to use metal oxide resistors in DC-coupled output stages like this.Just like C-type (high-HFE) transistors should be avoided too.There's a risk of self-oscillation. If you fitted metal oxide resistors in the amplifier circuit where B&O didn't, I suggest you replace them back to standard 5% or 10% carbon resistors, and it'll be fine. And there's no need for "increased stability" or sharp tolerance components anywhere in that circuit, - the trimmer adjusts and compensates for any off-tolerance components anyways and it's in itself a fairly high-tolerance component. Output stage transistors rarely go bad without dragging more down with them.Check the driver, the emitter resistors etc. and don't forget the remaining output stage transistors as they will haveexperienced overloads and may easily have developed scars that will show sooner or later. Martin
Thanks Martin for the suggestions,
I didn't use metal oxide resistors as you suggested but the Vishay CMF 1% metal film resistors which is different from metal oxide if i'm reading this correctly unless you tell me that, per B&O design, only carbon resistors should be used. Same goes for the signal transistors where I'm using only "B" type.
But the above statements leave me to think you know who was my teacher!
The only remaining parts to check / change are the emitter resistors which I don't have the right tools to check and the diodes... But if I'm decomposing the symptoms correctly, diode problem would make more sense than emitter resistors because the problem wouldn't self correct when removing the -35V rail, right?
Thanks for replying to my help
Charles
Craig: Jean You don't mention any diodes in your list above, when I blasted an output channel on a BM4400 Martin recommended replacement of the following components........ I suggest: Replace:0IC200, 0IC201, 0IC202, 0IC203TR220R277R276C238At least check (ideally replace):0TR200TR215TR216TR217TR219D204D205R271R272R278R279Martin
Hi Graig,
I read your post with great attention before posting and actually printed the above list as a resource. The only parts I didn't change or check are the diodes and emitter resistors.
But to me, it really sounds like a leaking diode but I can also be wrong...
BTW, good job on your restoration, another one saved from the trash bin LOL
Here are the pictures of the BM4400 heatsink to feed the conversation. As you can clearly see the gap between the heatsink fins touching the amplifier back plate, were the output transistors sit, and the marble surface. I would estimate this gap about 0.010" which is impossible for the mounting screws to catch up...
Another picture now of the back plate, you can clearly see that out of the 6 fins, only 3-1/2 are touching...
A little bit of sanding proves it...
Why not putting all the chances on my side, all heatsinks were flat sanded and will apply a try bit of thermal grease between the heatsink and the back plate. This should improve quite a bit heat transfer from the back plate and the cooling fins!
I'm back with fresh news
After the lapping of the heatsinks contact fins (above picture), I applied a thin coat of thermal paste over them and assembled back to the backplate as shown. By doing this, I'm confident that I maximized the performance of the heatsinks otherwise it doesn't hurt anyways.
Installed the output trannies using silicon pads instead of the mica / thermal paste combo. There's a lot of discussion around pro and cons of both. I prefer the pads because the work is cleaner, they are easier to install and do not degrade in time unlike thermal grease. Not to mention that there could be air bubbles between the mica and the silicone, which greatly affects negatively the performance of the thermal transfer.
I also used a thermal pad but thicker this time to transfer the heat from the backplate to the thermal protection thermistor. I did that because I still can't explain why 2 output transistors died while all other components in the upstream circuits were perfectly good. I can only explain this by the fact that they were badly cooled and died by over heating. I'm hoping by adding this extra precaution that the output transistors will have a very long life.
Yes, the picture represents a barely recognizable BM4400. There are still other steps to accomplish before the final assembly however at this stage, I focussed my energies to troubleshoot the signal path and output stage circuits. I made the first adjustments and slowly fired the skeleton; so far so good. Plugged my 5" test speakers and my laptop thru the tape input, OMG! beautiful undistorted sound coming out of the speakers already. The amp is already showing lots of character!
You will notice a lot of red into the pre-amp circuit, I decided to use mostly WIMA film capacitors from 5uF and lower instead of electrolytics, too many advantages to ignore them especially right onto the signal path.
As I said earlier, I didn't find failed parts (excluding capacitors...) in the output stage other than the two right channel TIP141 and this is a mystery to me.
Now, I would like to test the protection circuit to make sure it is working.
Is there a way to test it without making a physical short in the outputs?
Thank you!
Very impressive work Jean....I particularly like the silicone pads alternative to the rather messy thermal paste I generally use.
Craig
Thanks Graig,
The thermal pads make this part a little easier at least but you know how involving an entire restoration is...
Keep the good work Graig!
Not sure if this is very wise......but you may try holding a soldering iron close to R40 (PTC) to raise its temperature, this should energise the fault relay.
Craig: Jean Not sure if this is very wise......but you may try holding a soldering iron close to R40 (PTC) to raise its temperature, this should energise the fault relay. Craig
Hi Craig,
Actually it is wise but more effective with an hair dryer, thanks for the tip! The relay cuts after a while but the overload lamp does not lit. If I looking at the schematic carefully, I do not see any feedback from the thermistor to the overload lamp circuit and that seems normal. I think the overload lamp will only lit in case of high load, shorted output or distortion in the output stage... but this part is not 100% clear to me.
I'm almost done at this point, I just found a little too much of crosstalk on the left channel, to my taste, from the FM to other inputs... when the volume is maxed out.
Thats my understanding of how the fault circuit works too, so that seems good to me....as for the cross talk it will require guidance from Martin to solve ;~)
Craig: Jean Thats my understanding of how the fault circuit works too, so that seems good to me....as for the cross talk it will require guidance from Martin to solve ;~) Craig
Thanks Craig for the confirmation. I will try now with two 4 Ohm resistors in parallel on each channel and see if I can trigger that overload lamp
As for the crosstalk, it seem to happen around the TAPE 2 switch; there's a lot of traffic around that spot but I need to investigate a little more to find the source of it.
I'll let you know what I find...!
OK, here is the status,
After further investigation, not much I can do about the crosstalk. I will have this symptom only when there is one source playing while another source is active and it is audible only when the volume is maxed out. As far I'm concern, this is resolved and totally normal for the technology of that era.
The other thing that bugs me the most is a hum/buzz I'm hearing from 3/4 to Max of the volume scale only from the rear inputs (Phono, Tape 1 and Tape 2). Phono and Tape 1 are populated and I shorted the Tape 2 in/out. However the FM is just fine so that eliminates the tone control and output stage board. To test the FM input, I just turned down both volume trimmers (muted) then I cranked the volume to max.
So I'm concluding that something in common is wrong with the pre-amp board. Should I add a ground strap between the pre-amp board ground circuit to the chassis? Possibly a problem with 4TR1 transistor? C1 was changed BTW. Anywhere else I should look?
Hi everyone,
Project is now completed; I now have a brand new restored Beomaster 4400.
I want to thank everyone who helped me into this involving but exciting project. I learned something new, again…
A special thanks to Dillen who provided replacement parts and tips here and there; your help is immensely appreciated and it is amazing to see such knowledge for units produced more than 40 years ago. My hats off to you Martin!
This unit was purchased from Ebay; luckily it came in good shape and the cosmetic condition was outstanding!
A quick visual inspection revealed a blown power supply capacitor (infamous ROE red cap); I decided to start the restoration right away since there was no other sign damage and the inside of the unit looked like if it had very low hours.
The only mechanical issue was a broken finger of the FM tuner slider. I ordered the piece from Dillon, however, it was lightly different from the one I had so I repaired the broken finger making another other one from thin copper and epoxied to the existing one. One of the fingers most likely broke off because of the tuner pulleys are stiffening quite a bit over time to the point of becoming very hard to move. Silicon lubricant on each pulley pivot, including the tuner encoder, made the whole mechanism very smooth to operate.
The electronic part went very well. The whole preamp and tone control circuits benefitted of film capacitors only. I believe that film capacitors are better in every aspect to electrolytics and in forty years from now; only few new electrolytics will be required for third life. The only mistake I did was to reverse the polarity of the capacitor C15. This was causing to mute completely the FM when the AFC switch was activated. There were two blown output stage TIP141 on the right channel and were replaced with Motorola’s but all the other checked well with my tester.
I bought a variac for this project and glad I did it. When I first powered the unit everything seemed good! Sound was coming out from both channels, however, I could notice by ears some distortion on the left channel.
The scope came to the rescue; plugged both 4 ohm dummy resistors and I could notice a lower output from the left compared to the right channel… OK suspicious, raised the volume and I could see the left channel clipping around 50% of the volume scale and the left channel heat sink was cooking. All of sudden, smoke came out from one emitter resistors… too late. Checked the output transistors on the suspicious channel and they were all gone… dead… kaput! Couldn’t believe it! I ended ordering all new trannies and checked all the components on that channel; all good… bizarre!
After installing the new transistors and changed the smoked emitter resistor, I gave it another try and bingo! Nice identical sine waves on both channels so it’s time for the real benchmark. Adjusted the quiescent current and checked again after an hour, let’s go! Plugged the 4 ohm dummy loads, setup my THD meter and shot a 1kHz sine wave… and cranked up the volume until the overload lamp came on, backed just enough to clear the alarm, then read the THD value… 0.02% distortion WOW!!!!!!!!
Trust me this is quite a heavy load on the outputs and having such low distortion level is quite surprising and comforting at the same time!
I did some other tests but one last before closing the hood, I measured the DC offset and saw a nice -0.1mV on both channels, cool! It’s good to go!
Made one more tour before closing the cover and dammit, FM is no longer working at all either from the presets or the tuner. I was suspecting at first a defective TCA420A chip but to finally reveal broken jumper leads ( 4 out of 6 )from the IF board to the tuner box.
What I’m retaining from the few projects I did so far it that having the right tools is essential to make things right, willing is not good enough. The 4400 will be matched to my recently restored Beovox M70s and so much hard work was put into these that it would be a shame cutting corners on the amplifier work. Do yourself a favour and get the right tools when restoring a unit.
Now… Showtime!
Left and right channels sine waves sitting on top of each other perfectly!
Hand crafted tuner slider finger made of what? Copper house piping!
Now epoxied to the slider, sharp edges smoothed, and shrink tubed!
Before closing the top cover!
Front
Front 2
Front 3
That really is a cracking job, you have one of the finest examples I have seen of a BM4400. Cosmetically it is in pretty much pristine condition, I have never seen a brand new one of course but its hard to imagine a new one could be much different to your machine right now.
About the film capacitors you have used, I follow Martins recommendation and replace Tantal with Tantal....after all the originals have lasted over forty years, I notice that Rudy of Beolover fame also favours these new film caps too. I quite like the look of the squared off caps but I'm not knowledgeable enough to make an informed decision as to which are best suited to our applications......I'm happy with the tantals right now. Martins thoughts would be interesting..........
Congratulations, what a wonderful looking piece of kit.
Regards Graham
Craig: Jean That really is a cracking job, you have one of the finest examples I have seen of a BM4400. Cosmetically it is in pretty much pristine condition, I have never seen a brand new one of course but its hard to imagine a new one could be much different to your machine right now. About the film capacitors you have used, I follow Martins recommendation and replace Tantal with Tantal....after all the originals have lasted over forty years, I notice that Rudy of Beolover fame also favours these new film caps too. I quite like the look of the squared off caps but I'm not knowledgeable enough to make an informed decision as to which are best suited to our applications......I'm happy with the tantals right now. Martins thoughts would be interesting..........
Yes, I was very lucky to get this unit. They don’t come by that often here so you get what you can get…
You can’t go wrong if you follow Martin’s advice; tantalum vs film capacitor is not only about longevity but also about other considerations.
When I’m selecting components, my goal is to maintain the signal path integrity as much as possible. Depending on the available space, characteristics, voltages, I'm trying to use film capacitors first, then tantals then electos, etc… whenever has the least undesirable effect on the signal.
Another aspect to consider is, 45 years ago, a film capacitor was much bigger than today… The only place I replaced the tantals by films are the loudness and phono preamp circuits, all the others were left to tantals. So really, it’s not a black or white situation but more using components based on their applications.
However replacing the tantals to films wasn’t the biggest win IMO, I feel that the biggest win was to replace the small electros to films and there are plenty of them into the preamp and tone control circuits.
But my point wasn’t to debate the film vs tantal vs electros, my point was to elevate the importance to work with the right tools. I consider myself a newbee into vintage restoration but I love it so much, few years ago I bought myself the right tools to do things properly. As far I’m concerned, a repair and a restoration are two different things BY FAR!
Cheers to all!
Oh Yes!
Beovox M70 = Check!
Beomaster 4400 = Check!
Beogram 4002 = Next...
I forgot to mention a special thanks to my teacher; Menahem Yachad from Condor Audio.
http://www.condoraudio.com
Your help was much appreciated Sir!
Thanks a bunch
I have been following your thread quietly, and being very impressed how you solved things now, that 2 years ago, you didn't have any clue about.
The best reward to a teacher, is a student who succeeds.
G-d bless you for much success!
Menahem