Telefunken Tape Machine Level Matching

[Dr_J]

Hi Everybody,

I need some help here.

I'm building a little break-out box for connecting my old Telefunken tape machine to my interface (DIGI, +4dBu).

The Telefunken manual says:
INPUT: 0.15mV@2k (or [email protected] Turntable/phono)
OUTPUT: 1.55V@10k (this could be the old +6dBu German standard!?)

I think the voltages are RMS values, aren't they?

I think the output level and impedance should be fine?  ???
What about the input level and impedance. Should I treat this?

THANXS And ROCK-ON!

 

[ruffrecords]

I think you are right; the output should be fine. The input spec seems more like a mic input level. Does it have a separate line input? Which model is it?

Cheers

Ian

 

[Dr_J]

Hey Ian,

great to hear from you!!

It's a Magnetophon 250.

It has 2 kinds of inputs as described.
Radio/Mic: 1.5mV@2k
Turntable/Phono/Tape: [email protected]

I really appreciate your support, as ever!
THANKS!


ROCK-ON!

 

[RuudNL]

The M250 is a 'consumer' tape recorder, so I suppose you can control the input (line) level when recording.

 

[Dr_J]

Hey RuudNL,

yes, it has an input level control.
Question is how to feed it the best way?

Do you have the full schematic?
I couldn't find it yet, could you make it available to me?


ROCK-ON!

 

[abbey road d enfer]

Hey RuudNL,

yes, it has a input level control.
Question is how to feed it the best way?

The phono input should work OK, however a better solution IMO is to make an attenuator with a 10k resistor in series and a 1k in shunt to ground, and feed the mic input.
This way, the preamp would see a much lower source impedance resulting in a better noise performance.

 

[Dr_J]

however a better solution IMO is to make an attenuator with a 10k resistor in series and a 1k in shunt to ground, and feed the mic input.

Great, that is what I was thinking of. Thanks!

It's the shunt resistor, which effectively sets the value of the impedance seen by the interface or
preamp output, isn't it?


ROCK-ON!

 

[abbey road d enfer]

It's the shunt resistor, which effectively sets the value of the impedance seen by the interface or
preamp output, isn't it?

Roughly, yes.

 

[ruffrecords]

The phono input should work OK, however a better solution IMO is to make an attenuator with a 10k resistor in series and a 1k in shunt to ground, and feed the mic input.
This way, the preamp would see a much lower source impedance resulting in a better noise performance.

The line level input seems to simply add a 2M2 resistor in series with the input. Looking at the schematic it appears to be a kind of virtual earth input with a 22K feedback resistor. Just wondering if the 1K to ground would simply increase the noise gain?

Cheers

Ian

 

[Dr_J]

Hey Ian,

Ok, what would be your plan?
Should I replace the 2M2?

Cheers!

 

[abbey road d enfer]

The line level input seems to simply add a 2M2 resistor in series with the input. Looking at the schematic it appears to be a kind of virtual earth input with a 22K feedback resistor. Just wondering if the 1K to ground would simply increase the noise gain?

Actually there are two NFB paths, one V-to-I that encompasses the 1st and second transistors, and one V-to-V that is global.
How these two balance in terms of noise is quite complex and would need either simulation or experimentation.

 

[abbey road d enfer]

Should I replace the 2M2?

Decreasing this 2.2Meg resistor would increase the actual gain to a point that would make it unusable.

 

[Dr_J]

Ok, I think I’ll go for the attentutor at first glance.
Frequency (low end) response shouldn’t be a problem, right!?

Ian, your plan? 

Cheers

 

[ruffrecords]

Ok, I think I’ll go for the attentutor at first glance.
Frequency (low end) response shouldn’t be a problem, right!
Ian, your plan? 

Cheers

Abbey is right, the two nested feedback loops do complicate things. I would try the attenuator first but I think 10 and 1K is not sufficient. The mic sensitivity is 0.15mV so 10K/1K would only make this 1.5mV. I think you need at least 10 times this so 100K and 1K would give you 150mV sensitivity, same as the existing line input.

Ideally I would bypass this stage altogether and go straight into the 10K pot. No noise problem and potentially infinite headroom.

LF response should not be an issue.

Cheers

Ian

 

[Dr_J]

Hi Ian,

thanks a lot!!

Which pot do you mean? R12? ... and by this bypassing T101, T102, T103??


--- Interesting!
I just got a service manual from Ruud (THANKS MAN!!  8))
It says 0.6V@4k7 for the OUTPUT...!?  ???
That's wired...  :-\

The schematic from Ruud show an attenuator at the output with 22k and a shunt of 4k7...
Maybe i'll have to open mine to see what i've got...
Should I change the output attenuator to a 10k shunt?

I guess I'll have to do some tests...


ROCK-ON!

 

[ruffrecords]

Hi Ian,

thanks a lot!!

Which pot do you mean? R12? ... and by this bypassing T101, T102, T103??

Yes, R12. You would need to disconnect the output from T103 or maybe use a switch to select line or mic?

Cheers

Ian

 

[Dr_J]

Hi,

I just did a quick first playback test with an old tape... the return level in PT is around -60db!! 
I think this points to the 4k7 attenuator...

Next input testing....  ;D

Ian, thanks for the hint!! 


ROCK-ON!

 

[abbey road d enfer]

I just did a quick first playback test with an old tape... the return level in PT is around -60db!! 
I think this points to the 4k7 attenuator...

I don't think so. The input Z of your interface is probably higher than 10k, which suggests an attenuation of only about 3dB.
I would say it's either a wrong connection or a faulty component. I would check the electrolytic capacitors.
I had a similar problem with a Revox A77, it turned out most of the 'lytics had lost their electrolyte, so what was left in terms of capacitance was a few hundred picofarads.

 

[Dr_J]

Hey Abbey,

thanks.

The input Z of your interface is probably higher than 10k, which suggests an attenuation of only about 3dB.

Mmmm....  not sure...

I would say it's either a wrong connection or a faulty component.

That was my first thought too...  ;D
But if no idea what to change... I'll think i'm right here.
:-[  Ring, Sleeve mixed up.... better always tripple check...

I would say it's either a wrong connection or a faulty component. I would check the electrolytic capacitors.
I had a similar problem with a Revox A77, it turned out most of the 'lytics had lost their electrolyte, so what was left in terms of capacitance was a few hundred picofarads.

Thank for the hint, I'll have a look at this when it comes to open the machine... 

ROCK-ON!

 

[Dr_J]

Hey Guys,

short update. I'm making some (slow) progress...

I opened the machine and it has a L-Pad (22k/4.7k) on output has shown in the schematic.

I did some tests feeding in 1k [email protected].

For a L-Pad on input i ended up with Ian's guess (100k/1k) which gives me 7.5mV input (-44dB).
I tried some higher attenuation values, but than i had to crack up the input gain too much which results in increasing noise. So I think it's better to hit it hotter...

For the output i set the level of the internal meter to read "0db" (Fader position "3").
I modified the existing L-Pad and changed 22k to 8.2k which results in a return of 1.26V - close enough.

I did some first frequency response test; not happy.
I just ordered a degausser - never heard of that before...

ROCK-ON!