Sign in   |  Join   |  Help

Beomaster 1900-2 T2903 dead after recap

rated by 0 users
Not Answered This post has 0 verified answers | 48 Replies | 0 Followers

Andreas
Not Ranked
Vienna
72 Posts
OFFLINE
Gold Member
Andreas posted on Tue, Dec 1 2020 12:35 PM

Hello, i wanted to rejuvenate my Beomaster 1900-2 Type 2903 which i own for a very long time. I replaced all my capacitors, rectifier, lamps and idle trimmers with Martins kit. I also changed the 2 big reservoir capacitors with two F&T Al-Elko 4700uF -10+30% 40v 

I took my sweet time and worked very carefully, double/tripple checked everything and checking polarity of capacitors. When plugging it to a power source the machine is just dead. There is no click, no light, nothing...

I again checked every solder joint but everything seems to be fine. Do i have to do anything else when swapping so many parts ? My BM worked fine before i just wanted to give it back some freshness in sound and to get rid of the humming noise on my left channel. I got the capacitor kit from Martin and from the 100uF there were only ones with 50v where as some original capacitors read 100uF 16v so i guess that is ok ? I of course used these for all the 100uF capacitors. There is also this big 2200uF radial capacitor in the kit for C92 which has arrows pointing in both direction, usually the arrow points to negative. I went with the dented end to be positive. Is that correct ? 

I uploaded two pictures on to my profile, somehow i cant insert it into my text here.

I really need to get this beauty back running, every help is very much appreciated 

 

All Replies

manfy
Top 500 Contributor
132 Posts
OFFLINE
Bronze Member
manfy replied on Sat, Dec 5 2020 10:39 AM

Andreas:

I hope everything seems to be ok now, i'm feeling very tempted to turn it on but will wait for your last word :)

NO !!!! Do NOT turn on the unit !! You will fry your new TR27! Crying

Explanations:

- you measured 2.8mV at the collector of TR27: The only plausible condition where this can happen (with 20V on the emitter) is if the 15V rail is shorted to GND. And that is a failure condition!
The backplane of TR27, which I call the heat sink of the transistor, is electrically connected to the collector. If this node is connected to chassis ground, i.e. the external heat sink for TR27, you're short-circuiting the 15V rail. The collector of TR27 MUST be insulated from the external heat sink and GND. That's what the mica sheet is for. 

You can test your board by not mounting TR27 on the heat sink. Since you only switch on the main switch without bringing the unit out of standby, this external heatsink is not needed. The idle current in standby should be only a few milliamps.

So, all you want to do right now is getting 15.7V on TP16. You don't want to run the whole unit yet.

"TP16 resistance to GND = goes up gradually starting 1k ohm 10k ohm upwards "
That sounds ok to me. You're charging C97 when you measure between these 2 points and that explains the rising 1-10k value. What do you measure on the donor unit between these 2 points?

Your transistor tests and voltages given are insufficient; you also must measure C-E; you should read infinity in forward and reverse direction. Even if it does, it is no guarantee taht the transistor is good. The C-E junction could be permanently open even if B-C and B-E read fine.
But that's nothing to worry about right now. 

"Anyway i soldered the new TR27 in and put everything back together and there is now no steady connection to GND between the backplate of TR27, heatsink and GND"
What do you mean by "steady connection"? An electrical connection? If yes, that would be bad as explained above.
Oops, sorry! I misread your post. You said "NO steady connection" and that is fine!

Anyways, since you have a donor unit now you should compare your resistance measurements between these 2 units. If conditions between these 2 units differ (e.g. TR27 mounted on heat sink on one unit but desoldered on the other) then pls says so.
Pls measure TP14/15/16/17 to GND (and use the same probe polarity in both units!)
All these test points are in and around the 15V circuit and the results can tell us something about the overall condition. That will help us decide if and when it is safe to do an actal power on test.

Thks.

Andreas
Not Ranked
Vienna
72 Posts
OFFLINE
Gold Member

Yes i will wait, thank you :)

Andreas
Not Ranked
Vienna
72 Posts
OFFLINE
Gold Member

The new TR27 is now properly installed with mica sheet and thermal paste. There is no connection between TR27 collector with the black heatsink when testing with multimeter in beep mode. When in diode mode the voltage does go up till 1v and then OL 

Measurement TP16-GND on the donor unit is same 1-10k ohm

When testing all the transistors i also measured reverse direction and these were all infinity/OL, sorry that i didnt mention this. 

forget the ''steady connection'' that was bad explaining from my side. As mention above, it should be mounted properly now Smile

Should i still take TR27 off the heat sink and then turn the unit on and measure voltage on TP16 ?

manfy
Top 500 Contributor
132 Posts
OFFLINE
Bronze Member
manfy replied on Sat, Dec 5 2020 11:56 AM

No, if TR27 is insulated now against GND then leave it as it is.

But please do the comparative resistance measurements with the donor unit on all TP test points mentioned above, before powering on. Maybe also compare the zener diodes without removing them. That can tell us quite a bit.

Andreas
Not Ranked
Vienna
72 Posts
OFFLINE
Gold Member

Repair unit

TP14 - 2-8k ohm

TP15 - 14.7 ohm

TP16 - 1.6-10k ohm  

 

Defect donor unit without TR27 installed, not sure if we can trust these measurements

TP14 - 2-8k ohm

TP15 - 0.4k ohm

TP16 - 1-10k ohm

Andreas
Not Ranked
Vienna
72 Posts
OFFLINE
Gold Member

Diode test on board 

Repair unit

D51 - 0,77v rising,   reverse - 0,69v

D52 - 0,77v - OL,    reverse - 0,66

 

Defect donor unit without TR27 installed, not sure if we can trust these measurements

D51 - 0,77v rising ,  reverse - 0,66v

D52 - 0,8-OL , reverse - 0,7v

manfy
Top 500 Contributor
132 Posts
OFFLINE
Bronze Member
manfy replied on Sat, Dec 5 2020 1:21 PM

Looks ok to me, EXCEPT: 14ohms on TP15?
That seems too low. That's over one amp once you bring the unit out of standby and the BD136 (TR31) can only take 1.5A

I think there's something wrong and worth investigating. Disconnect PCB7 so that you get rid of the low ohm light bulbs in parallel (IL1/2/3) and measure again.

Andreas
Not Ranked
Vienna
72 Posts
OFFLINE
Gold Member

PCB7 is the one with the little board with three bulbs ? 

TP15 without PCB7 at 26,3 ohm          so on TR31 there should be around 0,5A right ?

manfy
Top 500 Contributor
132 Posts
OFFLINE
Bronze Member
manfy replied on Sat, Dec 5 2020 3:48 PM

Well, yeah. But I still don't like it.

I crunched some numbers:
The manual claims your unit has an idle consumption of 25W. At 240V that makes some 100mA in the mains line. The power amp will probably need 200mA idle current at 31VDC, which translates into 20+ mA in the mains line. So that leaves 80mA for the 15V transformer; roughly 1.2A at 15VDC.

You're measuring 14 ohms of pure ohmic resistance in normal state; that means >1A. In the circuit diagram I see no immediate justification for such a low ohm value. Once you power on, all transistors on that rail will come on - and that's what draws the real idle current. Additionally on power on all caps will be charged and you see a rush-in current that's 3-5 times above the static value -- at least for the first 50-100msec.
That goes beyond the limit of the power supply. No designer in his right mind would overlook this. If anything, B&O designers would choose a bigger safety margin than no-name makers.

The 400 ohms on TP15 that you measured on the donor unit would make me feel much better. But it's your Beomaster and your choice: Do you prefer to see smoke (and I mean that literally) or would you prefer to do more offline testing first?
The problem with "smoke" is not really just another dead component, but what kind of damage this might do to other parts of the BM1900.

Andreas
Not Ranked
Vienna
72 Posts
OFFLINE
Gold Member

I see, of course i wouldnt want anything to blow. I would like to tell you a big thank you for going through this with me step by step. If possible i would like to send you a good austrian wine for your effort :)

Do you have an idea which components i could check next ? I am thinking if there could maybe be a blown fuseable resistor around TP15.

I also discovered a resistor soldered on the bottom board between TR31 Base and TR30 Emitter on the donor unit. I guess this is a newer updated BM. Maybe this has something to do with the different readings from both units ? 

manfy
Top 500 Contributor
132 Posts
OFFLINE
Bronze Member
manfy replied on Sat, Dec 5 2020 5:43 PM

Andreas:

I see, of course i wouldnt want anything to blow. I would like to tell you a big thank you for going through this with me step by step. If possible i would like to send you a good austrian wine for your effort :)

Well, I'd have no problems with that -- but let's get the Beomaster running first. Smile

Andreas:

I am thinking if there could maybe be a blown fuseable resistor around TP15.

A fusible resistor? In the 1970s? In a consumer product? Unsure I'd say the chances of finding one of those are rather slim...somewhere in the range of 0.00000x%...

Andreas:

I also discovered a resistor soldered on the bottom board between TR31 Base and TR30 Emitter on the donor unit. I guess this is a newer updated BM. Maybe this has something to do with the different readings from both units ? 

No. That would have no effect on the resistance between TP15 and GND. Forget about any updates you might find on the possibly newer donor board. Your board did work before recapping, so the basic design is sound. Don't even think of copying some of those "updates" in newer boards as long as you don't know exactly what they do.
The only unknown now is the potential errors that may have occured during the recapping process. As long as you haven't applied full power to the boards we can assume that all components that are untouched are working normally. And we're trying to keep it that way by not doing too many of those "let's see what happens if I do this"-tests. Wink

 

 

Andreas
Not Ranked
Vienna
72 Posts
OFFLINE
Gold Member

Since the stand by light never lit up, i hope the whole unit was never powered up in the first place. 

manfy
Top 500 Contributor
132 Posts
OFFLINE
Bronze Member
manfy replied on Sun, Dec 6 2020 9:30 AM

Andreas:

Since the stand by light never lit up, i hope the whole unit was never powered up in the first place. 

Yes, exactly.

I looked at the circuit diagrams again, trying to figure out the safest way to test the boards. The design is rather confusing and there's no way to measure every node on the 15V rail and I fear we may have to do one of those "let's see what happens" tests and just hope for the best.

Here's what I would do if I had that unit on my bench:

1) First one final resistance test on TP15: unplug the tone control board 4 and board 7 and see how much the TP15 to GND resistance increases. I guess it won't change much, but if it does, we know where to look next.

2) If resistance did not change much, I'd be prepared to prep the unit for a first power-on -- but only for the 15V side because I don't want to see smoke from several corners at the same time.
- first I'd disconnect the +/-31V supply to the power amp. My reasoning behind it: If there's no power supplied to that end then nothing can blow up if something is wrong. Unfortunately we're also losing the -12V supply, which is required for several functions on other boards but I think that shouldn't be a problem. All we want to see is that 15V on TP15 is available when we bring the unit out of standby and if the 2A fuse F1 doesn't blow there's a good chance that the circuitry on the 15V rail is fine.
Looking at the diagram I'd say it's easiest to unplug or desolder pin 8a and 6a from board 8. That's the AC side of the +/-31V supply.
Do not simply unplug the relay RL1 because it is needed to activate the 15V on TP15.
- I'd keep the tone control board 4 connected to board 2, but the light bulb PCB 7 disconnected in order to get the smallest possible current on this rail. As a result we may be unable to put the unit back into standby with the OFF-sensor button, but that's not a problem because we only see this as a first "smoke"-test for the 15V supply.

3) Now I'd do the first test of the standby voltage on TP16.
- clip your voltmeter to TP16 and GND
- plug in the power cord
- switch on the mains switch and keep you finger on the button. If you see smoke, switch it off quickly.
I do not expect any smoke at this stage, but well, everything's possible.
If the fuse didn't blow and if you read 15.7V +/-1.5V then the 15V regulator is fine and the standby LED should be on. So, TR27/28/29 and the zener diodes are fine (even though your diode test measurements above don't make much sense).
If 15V at TP16 did not come on, we'll have to start measuring.

4) If the voltage on TP16 is fine, we can start the real smoke-test: bringing the unit out of standby.
- when doing so, keep your finger on the mains switch, so that you can off the unit if necessary
- you should be able to bring the unit out of standby by pressing any of the sensor or preset buttons
- monitor the voltmeter to see if the load causes the 15V supply to drop. Normally it shouldn't.
- if all looks good, keep the unit running for a minute, just to see whether something overheats
- then switch off and unplug the unit

5) connect the illuminator PCB7
- clip your multimeter to TP15 and GND (that's the switched 15V rail)
- plug in the cord and switch on mains
- bring the unit out of standby
- now the lights should come on and if all ok, we have successfully finished the first power on test

Once this is done successfully, we talk about the next steps. Good luck!

PS: It goes without saying that the speakers are disconnected for all these initial tests! We don't want to potentially damage perfectly good speakers!

manfy
Top 500 Contributor
132 Posts
OFFLINE
Bronze Member
manfy replied on Sun, Dec 6 2020 11:43 AM

Eureka!

I found the reason for your seemingly low 26 ohms on TP15:

The volume indicator board 6 sits  right on the 15V rail and provides a direct path to GND. Those 2 bulbs are spec'ed at 12V/1W, i.e. 144ohms. That is the resistance at operating point. Let's assume a cold resistance of approx 10%, i.e. 2x14 = 28ohms.
Now your measurement of 26.x ohms at TP15 makes sense and you don't need to worry about too much smoke.

In light of this new discovery, I'd say you can skip the step of detaching the illumination PCB7.
Nevertheless, I'd disconnect +/-31V for now as described above. It's just a matter of "better safe than sorry"!

And now you also know what's broken on your donor board: a bunch of light bulbs, including but not limited to 6IL1 and/or 6IL2. Big Smile

Andreas
Not Ranked
Vienna
72 Posts
OFFLINE
Gold Member

TP15 without tone control and PCB7 - 26,4 ohm

TP 16 - 14,36v  on stand by 

TP 16 - 14,35 out of stand by 

TP15 + PCB7 - 14,36v out of stand by,  volume lights and PCB7 lights are on

the others on PCB5 are not lighting up

Everything seems stable and relay is switching

Wow, that was nerve racking... :D

 

haha ok i see, yea the bulbs on the donor unit are pretty much gone. With all the knowledge you gave me, i might take an attempt to repair that one aswell. Thats very good news that the readings of my BM finally make sense :))

No speakers were harmed in this process ;)

Cheers Beer

 

 

Page 3 of 4 (49 items) < Previous 1 2 3 4 Next > | RSS
Beoworld Security Certificate

SSL