• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Tandberg TR2045 recap??

pari57

pari57

Member
Hello,

I've already read the posts about the Tandbergs in the forum, but there are still some questions I have.
Maybe they are not significant for Tandberg alone and also apply to other brands, but that, I don't know.

so
here I go.

In one of the posts was mentioned. when exchanging caps, keep capacity and voltage as they are best don't move more than 1 step away from original value.

this sound logic.
But what if the specific cap can't be found?
to make it more real take this example
On the TR 2045 the 25 volt power line has 2 Plessey caps 100uf 40V beautiful green

I am sure the thunder & lightning sounds coming from my speakers in quiescent position indicate that I should change those (among with others of course)

But Tandberg must have had a reason to use a 40 volt part instead of the more usual 35 volt or 50 volt parts

So what to do here. 100uf 40 v is not available.

so where do we go? keep the voltage and change the value or change the voltage and keep the value.
and if so what way up or down.

e.g.

Choosing 35volt will overload the cap most of the time, even if it can have 35 + 20% = 42 Volt
On the other hand will it cut the max voltage at 35 volt instead of 40v before and even-though the regulated voltage is 25 volt. far lower than the caps voltage
Danger here is that the cap might wear out more quickly. short lifespan

Choosing 50 volt gives the cap a more relaxed life. a long life span, but it might interfere with the reaction speed of the system. loading a 50 volt cap takes more time than a 40 volt (think in light speed not human speed)

does it affect the rest of the system?

The same goes for the 2 main caps for the power amp line. Elna 10000uf 50V black neg.35x70mm
the modern caps are much smaller 30 x40mm and probably not as robust as the old ones.

Another thing I wondered abut is the variation in cap voltages throughout the assy's.

on the input board C507/508 are only 6,3v but C515/516 are 63v ???? and in between we have
C501/502 at 35v and C509/510 at 25v

so where is the logic here.
especially why the 10uf @63v make me wonder. btw the are spread all over the machine like that.

As you can see some questions that I would liked to have answered.

when there is some one around who can ? I would be grateful

regards

Paul
 
Register to hide this ad
pari57 said:
But Tandberg must have had a reason to use a 40 volt part instead of the more usual 35 volt or 50 volt parts
...
Choosing 50 volt gives the cap a more relaxed life. a long life span, but it might interfere with the reaction speed of the system. loading a 50 volt cap takes more time than a 40 volt (think in light speed not human speed)

does it affect the rest of the system?
Click to expand...
The voltage rating of a capacitor will affect its physical size, and a higher voltage cap often has a lower ESR than an equivalent lower voltage cap -- which is good -- but it will not affect "the reaction speed of the system". The time it takes to charge a capacitor is determined by its internal resistance and capacitance, not its voltage rating (per se).

Often, a company will choose a capacitor because it's available at a good price, not because of a particular engineering reason. Tandberg might simply have got a good deal on 40 volt (and 63 volt, and so on) caps. If there was an engineering reason for choosing 40 volt (or whatever) caps, it's almost certainly related to fitting the caps within the physical space rather than any electrical property.

Modern caps tend to be smaller than vintage caps for a given voltage rating. Therefore, you should be able to increase the voltage rating without running into space issues. Modern caps also tend to be very reliable and of excellent construction.
 
DaveVoorhis said:
Therefore, you should be able to increase the voltage rating without running into space issues. Modern caps also tend to be very reliable and of excellent construction.
Click to expand...

thanks for the reaction Dave,

maybe I have put you on the wrong foot here, that space is the problem, but that is not the issue.
when you know the TR2045 machine you will know that there plenty of room to put any cap where you want it.
even if one would go way up higher in values.
No the issue is

can I replace the 100uF 40v cap with a 100uF 50 volt or 100uF 35 volt cap
or put in a 40v 120uF or 40v 150uF or 40v 220uF without significantly changing the sound quality of the machine.

It is known that bad caps produce bad sound. but the wrong caps do as well.
Especially in high end machines.
 
If there's room, the higher the voltage rating of the capacitor, the better. Higher voltage ratings will not compromise the sound, and their (often) lower ESR may improve filtering.

For filter capacitors, it's generally safest to stick to the original capacity, as the inrush current draw of higher-capacity capacitors may damage rectifiers and put excess load on the power transformer, but higher values won't compromise sound. Other caps should be the same capacitance value or fairly close -- generally slightly higher is safe, lower is to be avoided -- to preserve the original sound.

Therefore, I would recommend the 100uf 50v capacitors over the 35v.

Note that "thunder and lightning" sounds are not necessarily due to capacitor problems. It could also be a result of failing semiconductors or poor connections.
 
Last edited:
thanks Dave


I was checking out the complete list and I will take the 50v versions.

Then there is the issue on the 10uF 63v caps. they are all in the signal line.
I guess they are coupling the different segments

I cannot get the Elna silmic 63v I can only get 50v.
I live in Germany and Mouser has only Elna 25v 50v and 100v and 100V only the 100uF and more

Since these are (de-)coupling caps I could image 50v would be good here
but being strait in the signal line I want the best there is; so Elna

any ideas here? can I take the 50v

about the bad semiconductors I have checked that out.
the connections will be taken care of as well.
I want to have all the stuff together before I start taking it apart.

The machine is working great, beside the fireplace and T&L sounds in the background now and then.
At the moment I am listening to Phil Collins - no son of mine on the radio.

every now and then one really thinks, don't do anything, the machine is okay, even after 36 years (prod date kw48-81)
But than lighting strikes and a the man knows "a man has to do what a man has to do".

regards

Paul
 
For coupling caps, 50v should be fine but you can also use 10uf film caps which are available in higher voltages and may be lower distortion than electrolytics.

How did you check the semiconductors?
 
I did a quick check with my scope.
Beside the distortions everything is in perfect order.
The power-amp line-voltages are a little to high 40.4v - 40.7 instead of 40.0v and they fluctuate some.
even the 25v is good and stable but the c905 1000uF is really getting hot.
I even found an exact same part for this one. It is on its way.

All the voltage values at the transistor outputs of the several transistors are at value or very close to the
given values. 17mV on R751/R752 is where it should be.!!
If there is really a problem with any semicon, I can only find out after I first eliminate the most obvious disturbances
and that is mostly the elec caps.
lets be fair, after 36 years either they must have to be worn out from working or lame from hanging around and do nothing.
anyway neither of that is good for another decade or two of listening pleasure.

BTW I am not looking forward to see that any of the semicons really is faulty.
They are very rare and even-though there are some replacements around,
it can be very tricky.
Find some that turn out to be fakes,
or find some and they charge you for it like it is pure gold.
and replace them with different ones will for sure change the sound.

to be on the safe side
about the small styroflex caps, should they be replaced, or the tantals ?
I guess the MKT's will be okay ?

regards

Paul
 
Caps getting hot are generally an indication they're bad.

Conventional electrolytic caps of any age, and vintage paper caps and some paper-in-oil caps -- "Black Beauty" or "Bumblebee" caps and the like -- are failure-prone and should generally be replaced on sight. Tantalum caps are generally fine as long as not subject to more than about 50% of their rated voltage[1]. Other types of caps generally last indefinitely, as long as they haven't obviously failed outright by being open or shorted.

--
[1] Vintage solid-state Tektronix oscilloscopes used tantalum caps, often run at close to their rated voltage. They often fail. I know of one case where such a 'scope owner wondered what a lit red LED meant on a particular board, given it was normally invisible to the user. This caused some confusion -- there isn't a red LED on that board -- until the user realised it wasn't a red LED at all. It was a shorted tantalum cap glowing red hot!
 
The C905 is the very one AK member @jdurbin1 is constantly warning for it will short. You probably did read that, already.
It is probably a yellow Frako one in your case?
I would replace asap, before worrying about anything else. I see you have it on order already.

On one of my my tr2075 I found having the input pair of transistors in one of the channels having very different gain, although it was the next stage of transistors being culprit of bad sounds (next stages are different from tr2045)
So for funny crackling sounds, make sure if coming from one main amp or 2 main amps. It may be coming from the bad 1000 uF cap, after all.....
 
hi Gslikker, dutch he, ik ook maar wel in berlijn.

Yes I expect it will be that one.
but since the machine is that old I don't want to have any problems in the future.
as soon as I close it I want to feel good for some years.
I also just bought a TCD 330 again to go with it and I am already getting the vibes to hear them together.
can hardly wait
btw I got some HiLo speakers
yes yes from Wormerveer.
very very undervalued stuff

gegroet

Paul
 
some photos about an irregularity I noticed.
regarding dismantling the unit, the service manual mentions nothing about as screw in the tuner knob.

It only says pull off the knobs.

the tuning knob however is locked with as small Torx screw.
I used the Torxdriver T6x 50 it fits perfectly.

see the pictures.
 
Not an irregularity, but early vs. later production variant.

They used set screws on the earlier TR-20xx/series 2 units (2025, 2055, 2075 MkII), push-on/pull-off type later. Both work with the same D-shaped skinny control shafts and the outside of the knobs was identical other than the hole for the set screw on the early type.

All tuning knobs used set screw/round shaft.

That said, 2045 is series 3 (amber display models) and I would have expected the push-on type with plastic inserts. Possible they were replaced at some point with earlier type knobs, though - they're 100% interchangeable.

John
 
Okay the recap so far has worked, but there are some things that puzzle me immense.

Pre AMP & Power AMP Board Copy.jpg

turning the machine on the first time after reconnecting all wires,
it stayed quiet; extremely quiet. !!!!
so quiet, one would believe the machine was dead.
Not a sparkle, no sniff, not a sigh, nothing.

(I did had no inputs connected, but still !!!)
so I only had the amplifier section connected in that moment.

but I checked again and my volume was completely open. but not a sound.

Amazing how quiet this thing is.

So I turned down the volume and connected the radio section.

And there was the radio signal. Amazingly clear, with only a short copper wire as an antenna.
Just unbelievable.

Everything was checked and seemed to work perfect.
The machine was on test for more than 12 hours.

But then this morning after turning the machine on, before the relais came on, there was a short burst
in the speakers. just a fraction of a sec. Apart from that, everything was working fine

Turning it on/off gave no new burst.

Leaving it off for a few minutes and then turning it on gave the short burst again.

the second puzzle after some hours suddenly the right channel started giving some leak signal
sounding very weakly even without the volume closed.

I found out that the sound was slightly amplified when I tested contact 6 and 12 on U601 with the Voltmeter.
checking the voltage on these contacts, it ran down from 7,45 v to approx 6,5 v.

checking the signal with the scope I noticed a rumble on the right signal-path back to Q702 1 and 5
it turned out the voltage on 3 was floating around (-0,122 to - 0,788). (left channel Q701 was a steady -0,645v)

I have no cold-spray, so I used some pressed-air-dose (this is a little cold) and sprayed on Q702
and notice: suddenly the sound was gone. the rumble signal was gone. quiet as in heaven.

So my Questions to you clever guys :

Where did the short burst at turning-on coming from ?

and

What caused the rumble-signal in the right channel to appear out of nowhere
and why did it disappear after the air-blow?

And an other question :

It seems the machine had a shower with something; cola or some other sweat drink.
all the boards are covered with this sticky stuff.
I would like to get it of.

Could I wash the boards with water and soap ( maybe in the dishwasher,.. haha )
and dry them with a hairdryer (60-70 °C)?
Has anyone ever done that before?
I mean most of the stuff on the boards is waterproof

I am curious
regards

Paul
 
pari57 said:
Where did the short burst at turning-on coming from ?
Click to expand...
The posted schematic is too small to read and I haven't bothered to find another one, so I haven't checked to see if it has speaker relays, but if it does it might be that one of the relays is sticking closed. In the absence of (working) speaker relays or a soft-start circuit, many amplifiers will normally produce a pop -- a brief pulse -- before settling down to normal operation. If it has a soft-start circuit, it might be failing.
pari57 said:
What caused the rumble-signal in the right channel to appear out of nowhere
and why did it disappear after the air-blow?
Click to expand...
Sounds like Q702 is either defective or has a poor solder connection.
pari57 said:
Could I wash the boards with water and soap ( maybe in the dishwasher,.. haha )
and dry them with a hairdryer (60-70 °C)?
Has anyone ever done that before?
I mean most of the stuff on the boards is waterproof
Click to expand...
Cover up the stuff that isn't waterproof and wash with soap and dilute dish detergent, then allow it to dry thoroughly. A hair dryer will probably blow water around; leave it out in the sun a few days. Washing circuit boards with water and a gentle detergent is absolutely fine. I've done it many times and it's used in industry. Electronic components are almost invariably sealed and waterproof, but avoid getting water in pots, trimmers, variable caps, or in and/or on transformers and inductors with iron or laminated cores. Also, avoid getting water on printed panels, tuning dials, etc., as these sometimes have water-soluble printing, particularly on older gear.
 
to Dave

Thanks for the replay,

I'll check the Q702 again specially the print connections. one never knows.
I had a lose connection in the radio connection. hardly to see,
but as I touched the connection unwillingly, there sounded a bad crack.
as I checked it, it turned out 1 of the print connections had loosen.
after soldering it was perfect again.

yes the machine has speaker relays.
It is a bit of a worries, but so far there has been not any indication that it could really be bad.
Except for being pretty old: 36 years.

But I will probably take the machine apart once more, to clean of the sticky stuff.

btw how can I get the schematics bigger ??
I just uploaded the original 4000x2500 2,5mb
thinking that it will be visible

regards
 
pari57 said:
btw how can I get the schematics bigger ??
I just uploaded the original 4000x2500 2,5mb
Click to expand...
The forum software limits uploaded image size. Better to post the URL of the site where you got the schematic, so we can download it from the source and be sure the version we're looking at is the same as you're looking at.
 
Dave, schematic is here: http://sportsbil.com/tandberg/tr-2045-circ.PDF

Not the neatest scan in the world as far as getting stuff to fit on the page but it's legible.

Agree with checking the relay contacts & cleaning as needed... it's essentially impossible to get a transient in the speaker output unless one or both of the contacts have stuck, or unless possibly the relay was driven closed instantly at power on and even then it should miss any transient caused by DC offset if everything else was healthy, I would think.

Q702 is the differential pair at the input of the amp, I don't usually have problems with them but could be it has become noisy and thermally sensitive. Anything that pushes DC to that zero crossing circuit across the amp from input to output is going to be problematic.

John
 
update on TR2045 Recap


The nasty burst at start-up turned out to be a speaker-signal-related problem.

For those who have the schematics of these machines:

The relay K771 is powered through a set of transistors Q774 and Q775 putting up a level of 28 Volt for the Relay.

I found out that some speakers - connected to the machine - made the relay start clapping, were as others did not.
So the signal had come from one of these signal lines (not the standard power line), that are sending their signal to Q774 and Q775 via
Q771 Q772 and Q773.

Since the problem did not occur with the old caps still in place, I sensed the solution had to be in that direction
.
C773 is the first one to come to mind since it is directly involved in Q774
As I checked the values of the old and the new capacitor for C773 it turned out the
the old one had a value of 26,4 uF and the new one only 23,3 uF.
the old cap was an elna 22uF 35V 20%
the new one a Elna silmic II 22uF 35v <10%

putting the old cap back in place solved the problem !
no more clapping no more bursts.

It does make sense that a too good cap can sometimes be bad

the second problem ;
The noise in the right channel seems to have disappeared after I have re soldered the print connections of Q702.

so far everything looks good, goooooooood. (bruce almighty)

greetings
 
You must log in or register to reply here.
Back
Top Bottom