Help calibrating a Sony TC-399

Dan LB

New member
Hi All,

I'm about to embark on calibrating my Sony TC-399. I think I have all I need:

  • Calibration tape (equivalent of Sony J-19-F2 from Gennlab.com)
  • 2ch oscilloscope
  • DMM that will measure VAC and frequency
  • Blank tape (AGFA PEM 468 - I know it's old but I have quite a bit of it and it's in new condition and not exhibiting any signs of SSS)
  • Tone Generator (Logic Pro X outputting through an Audient iD22)
  • Service manual (see attachment)

I have never done a calibration before and having read the relevant section of the service manual (starts on p17 of the attached .pdf), I could use some help.
I have some basic electronics experience as well as having worked professionally in the audio industry for the last two decades.

The machine has been cleaned and demagnetised as of this evening and is in good mechanical order.
I think I know how to set the azimuth of the playback and record heads but when it comes to the playback level and record level adjustments I'm a little unsure. I haven't made any adjustments yet.
The first tone on the test tape I have is recorded at 185nW/m. The meters aren't calibrated so I won't go off them yet. What voltage should I be reading at the line out for this tone? The manual says "0.435V (-5dB) reading". Does this mean I adjust the trimmer for that output voltage and then calibrate the VU meter to read -5dB?

Maybe I just need to start at the start and work my way through everything.

I'd appreciate it if anyone could help guide me through the process.

Dan
 

Attachments

  • Sony TC-399 Service Manual.pdf
    2.4 MB · Views: 37
Hi Dan...Corrr! Been decades since I saw a Sony OR manual! Good innit?

Your first problem I am sure will be the DMM? Very few are accurate past a few hundred Hz. Yes, it might well measure F up to 100kHz or bettter AND say "true rms" but that does not mean it has a flat DC to 100kHz response on AC volts. You can check this with the PC since any DAW will generate perfectly regular sines at 100 Hz and up, just hook the meter to the AI and see the relative reading at 100, 1000 and 10,000 Hz.

You really need of course aN ac millivoltmeter and these can be found on ebay but their accuracy is dubious. You could calibrate one of course but there is a fairly simple solution if you have a bit of soldering knowledge?

Buy the biggest and best VU meter movement you can afford and build a simple voltmeter around it using an IC. I can find you a circuit. The principle is the meter is in a diode bridge (Germaniums preferred) and that is in a feedback loop with the IC so the result is very linear. The response can easily be up into RF!

From a quick look at the manual, you calibrate the internal VUs on P19. I would not touch the bias traps, only need that if a component is changed? (Sweetbeats and other up to date tape bods will i am sure chip in) .

Take copious high res pics and start a note book and record EVERYTHING!

Dave.
 
Hi Dan...Corrr! Been decades since I saw a Sony OR manual! Good innit?

Dave!! Great to see you here! Yep, same Dan LB as from the SOS forums :)

Your first problem I am sure will be the DMM? Very few are accurate past a few hundred Hz. Yes, it might well measure F up to 100kHz or bettter AND say "true rms" but that does not mean it has a flat DC to 100kHz response on AC volts. You can check this with the PC since any DAW will generate perfectly regular sines at 100 Hz and up, just hook the meter to the AI and see the relative reading at 100, 1000 and 10,000 Hz.

My DMM isn't of the true RMS variety however I do have a true RMS type on the way - sounds like it might not be up to the job either though :(

You really need of course aN ac millivoltmeter and these can be found on ebay but their accuracy is dubious. You could calibrate one of course but there is a fairly simple solution if you have a bit of soldering knowledge?

Buy the biggest and best VU meter movement you can afford and build a simple voltmeter around it using an IC. I can find you a circuit. The principle is the meter is in a diode bridge (Germaniums preferred) and that is in a feedback loop with the IC so the result is very linear. The response can easily be up into RF!

Yep, plenty of soldering knowledge but only basic electronics knowledge as you know - sounds like a simple little project though

From a quick look at the manual, you calibrate the internal VUs on P19. I would not touch the bias traps, only need that if a component is changed? (Sweetbeats and other up to date tape bods will i am sure chip in) .

Not too concerned with the VUs just yet as I want to get azimuth and playback levels sorted. Bias traps eh? I've replaced some of the electrolytic caps (the previous owner replaced the rest of them) and most of the transistors...... Might need a tweak.

Take copious high res pics and start a note book and record EVERYTHING!

I'm usually crap at remembering to do this but I'll endeavour to do so this time!!! :)

I'll post here as I go :thumbs up:


Dan
 
Hello again!

Yes, it's me Dan, like a bad penny. Found a circuit for a simple mVm. Not the one I first thought of but will do the job.

The attanuator resistors are odd values in some cases but you can make them up and check the result on the shiny new meter! At least DMMs are bang on accurate for resistance! In any case you only need good, not NPL, accuracy and at a few dB/voltage values. It is MUCH easier to work with a direct decibel scale than diddling numbers on a digital.

The op amp used is not as common as the TL072 but I am sure the latter will be fine. Battery power is a bit of a bind but at least you have no ripple/ground loops to worry about and a pair of decent alkaline PP3s will last ages IF you remember to switch it off!

I see the Sony manual gives the "over the hump" at 1kHz bias method? I preferred getting 10kHz just 1dB down from 1kHz but I am long out of the game and others here will no doubt have better advice.

Oh! And you really should buy a head demagger. Never bloody ends does it!

Dave.
 

Attachments

  • audio-mv-meter.pdf
    767.7 KB · Views: 17
Cheers Dave, I have a demagger already - a very nice NOS Han-D-Mag :guitar:

Am I right in setting the azimuth for the play head as my first step?
 
Cheers Dave, I have a demagger already - a very nice NOS Han-D-Mag :guitar:

Am I right in setting the azimuth for the play head as my first step?

After a de-mag I would think so, as I say been decades! I do remember however that you need to be careful with azimuth on a stereo machine (I was fixing mono Grundigs!) because you can get the peak multiple waves "out".

IIRC look at the sines on the scope and make the weeeest tweak to azimuth to get them bang in phase.

BTW, if I were in your position, once you have the machine aligned to perfection, dump the test tape onto a PC then run off a copy and use that for odd checks, once a week say depending on usage. Put the virgin test tape in a steel box.

Dave.
 
I started my calibration process today. After cleaning the tape path with IPA I spooled up the calibration tape and set the azimuth for the playback head using the scope in X-Y mode to monitor the output of both channels on the machine. Using the 10kHz and 12kHz tones I got it into pretty good alignment

20190823_133241(0).jpg

Great! So now onwards to the record head. I generated a sine wave from my mac using Logic and sent that to the line inputs on the machine and confirmed it was getting there by monitoring from 'source' on the TC399 - so far so good.

I then put a blank reel of tape onto the machine and put it into record - this time monitoring from 'tape' as the recording was being made, except the machine no longer records! Any ideas on what might have happened?
I put a reel of tape on then with some audio already recorded and tried to record over it but I'm just hearing the pre recorded audio when in record mode - as if the machine was in standard playback mode. :confused:

Dave you mentioned earlier that the bias might need adjusting since I've changed components. Could this be the problem?

The red record lights do come on when I put the machine into record.
 
Is the VU meter indicating a signal?

If you have a X10 probe for the scope stab that on the rec head pins and you should see bias, lots of it!

If you have bias you might just make out the signal modulation on the edges of the waveform but that might be easier to see with a heavy music signal.

If no bias that could be many things, a wire off the erase head since some designs used the EHead as the oscilator inductor.

Bias nut no signal, work through the rec chain with scope.
Has the machine been recording lately? If not it could be a tarnished rec/play switch. Blast with cleaner then clank it through rec/play mode several times.



I have not studied the map as yet but will do if the above checks produce no results.

Dave.
 
Is the VU meter indicating a signal?

They show a signal when I monitor the source and they show the signal of what is on the tape, but as there is nothing getting to tape in record mode they show no signal when recording while monitoring the playback head.


If you have a X10 probe for the scope stab that on the rec head pins and you should see bias, lots of it!

I don't have any probes for the scope but the guy I borrowed it from has. I'll get them from him on Monday.

If you have bias you might just make out the signal modulation on the edges of the waveform but that might be easier to see with a heavy music signal.

If no bias that could be many things, a wire off the erase head since some designs used the EHead as the oscilator inductor.

I'll have to wait til I get the probes.

Has the machine been recording lately?

Yes it was recording before I changed the transistors.
The ones I changed were:

Q101, Q201
Q102, Q202
Q103, Q203


Q104, Q204 - on the OSC board
Q105, Q205 - on the OSC board

Q106, Q206
Q107, Q207

Q109, Q209
Q110, Q210

Q113, Q213

Q302, Q303

The voltages around the transistors were good after the swap (going by the schematic)


Dan
 
Last edited:
They show a signal when I monitor the source and they show the signal of what is on the tape, but as there is nothing getting to tape in record mode they show no signal when recording and monitoring the playback head.




I don't have any probes for the scope but the guy I borrowed it from has. I'll get them from him on Monday.



I'll have to wait til I get the probes.



Yes it was recording before I changed the transistors.


Dan

Transistors!! You mentioned changing some caps not 'king transistors! Small signal TO92 types can have any of 3 pinouts!

Cannot recall Dan are you UK? If so I have some scope bits you can have including a probel I am sure. No longer have a scope.
 
Transistors!! You mentioned changing some caps not 'king transistors!
Ah but I did mention it :)
Bias traps eh? I've replaced some of the electrolytic caps (the previous owner replaced the rest of them) and most of the transistors...... Might need a tweak.

Small signal TO92 types can have any of 3 pinouts!
The ones I replaced were put in correctly observing the pinout

Cannot recall Dan are you UK? If so I have some scope bits you can have including a probel I am sure. No longer have a scope.

Very kind of you Dave (as always) but I'm in Ireland.
 
Ah but I did mention it :)



The ones I replaced were put in correctly observing the pinout



Very kind of you Dave (as always) but I'm in Ireland.

Ok so I missed that! (old, one eyed and poorly y'know) Ireland? No worse than my lad in France and I post him stuf all the time. If you want the bits Dan, post me a PM with an addy.

Time for some serious voltage checks methinks?

Dave.
 
If you want the bits Dan, post me a PM with an addy.

Dave, thank you so much for the offer, I appreciate it, I really do, but I don't have my own scope (I borrowed this one) so it would only be a waste of your time and effort.


Time for some serious voltage checks methinks?

I'll crack on with that so! :)
 
OK, here's what I've found...... The image below show's the voltages for each leg of the transistors that I've swapped out, according to the schematic, the measurement of the original transistor and the measurement of the replacement transistor

TC-399 transistor voltages.png

There are 4 that I don't have measurements for the original transistors that got replaced. Do you think the voltages measure within tolerance? I imagine they didn't need replacing in the first place! :facepalm:
The reason I changed them was because of reading about transistors failing in these models (TC 366, 377 and 399).

Also I do still physically have the original transistors.


What I found strange is that I have a few conflicting things going on - the schematic says that certain transistors are of a certain type (eg. it says Q104 is a 2SC458 but the electrical parts list in the service manual says Q104 is a 2SC1345 - which it physically was). The electrical parts list also has all the ones that I changed listed as 2SC1345 but some of them physically were C945. Unfortunately I didn't take note of where the C945s were removed from as I presumed the electrical parts list to be correct and proceeded to replace all of those listed as 2SC1345s with KSC1845s.

Also on the schematic it says the voltage at the collector of Q101 and Q201 should be 11V but on the actual PCB there is a print with an arrow showing it should be 8V (which it measures) so perhaps the schematic I have is from an earlier version? Maybe there was a revision? I really don't know. I'm taking what's written on the PCB to be accurate in that case.

Anyway, what do you reckon about the measurements that I've shown above......


Dan
 
Last edited:
OK! I'm back up and running! The machine is now recording again. There was a broken trace on the OSC board. Soldered in a jumper cable and all is well - nearly!

Recordings are lacking in treble somewhat - bias?

Going to clean the heads again now and use a different tape to confirm.

I'll post results of that shortly.
 
Just cleaned the heads and tried recording on a different tape. MUCH BETTER!!! :guitar:

I'll start my calibration process over again tomorrow. :)

Bias probably still needs a tweak I'd imagine....?
Should that be done before anything else?


Sleep now :yawn:
 
Just cleaned the heads and tried recording on a different tape. MUCH BETTER!!! :guitar:

I'll start my calibration process over again tomorrow. :)

Bias probably still needs a tweak I'd imagine....?
Should that be done before anything else?


Sleep now :yawn:

Hi Dan, Sorry I did not get back to you after that marathon voltge table! I had to go for my monthly eye check and the drops leave me blurry after. Better now.

I WAS going to say, all those values look ok except Q113/203 where you have a 2.5ish b/e voltage differential. Cannot be, Silicon transistors have Vbe of 0.6.to 0.7V. But, you are up and running so no matter.

Treble loss: Run the test tape and yes, set the azimuth. If that does not deliver on spec treble you might have a worn head.

Dave.
 
Back
Top