UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Tape (Audio), Cassette, Wire and Magnetic Disc Recorders and Players

Notices

Vintage Tape (Audio), Cassette, Wire and Magnetic Disc Recorders and Players Open-reel tape recorders, cassette recorders, 8-track players etc.

Closed Thread
 
Thread Tools
Old 19th Oct 2017, 10:36 pm   #1
Eichberger
Triode
 
Join Date: Apr 2017
Location: Winchester, Hampshire, UK.
Posts: 42
Default EQ circuit testing

Would anyone be able to interpret the following test results for me?

I am suspicious that there is something wrong with the EQ circuit in my Tandberg 12 so I've been doing a bit of testing with an oscilloscope. I input a 2kHz sine to the inputs and traced it through to the recording head.

There is an initial recording pre-amplifier section which looked fine. I then took a trace at the output of the recording EQ (just before the recording current trim pot) and got the trace attached.

The trace shows a pulse shape. Is this normal or does this show a problem with the EQ circuit? Was it wrong to expect to still see a sign wave?

Any input greatly appreciated.
Attached Thumbnails
Click image for larger version

Name:	IMG_0919.jpg
Views:	132
Size:	136.5 KB
ID:	151039  
Eichberger is offline  
Old 20th Oct 2017, 9:39 am   #2
lesmw0sec
Octode
 
Join Date: Jul 2009
Location: Carmel, Llannerchymedd, Anglesey, UK.
Posts: 1,498
Default Re: EQ circuit testing

A sine wave should still be a sine wave, regardless of any EQ. It would appear that there is some overloading occurring there - either due to a fault or too much signal being applied.
lesmw0sec is offline  
Old 20th Oct 2017, 10:55 am   #3
TIMTAPE
Octode
 
Join Date: Jul 2006
Location: Perth, Western Australia
Posts: 1,965
Default Re: EQ circuit testing

How well does it record and play back music/speech?
TIMTAPE is offline  
Old 20th Oct 2017, 5:44 pm   #4
Eichberger
Triode
 
Join Date: Apr 2017
Location: Winchester, Hampshire, UK.
Posts: 42
Default Re: EQ circuit testing

Quote:
Originally Posted by lesmw0sec View Post
A sine wave should still be a sine wave, regardless of any EQ. It would appear that there is some overloading occurring there - either due to a fault or too much signal being applied.
Yes, it was overloading. I'd somehow managed to turn the record level to max while moving the machine about! Getting much better results now but still rather puzzling.
Eichberger is offline  
Old 20th Oct 2017, 10:29 pm   #5
kevinaston1
Hexode
 
Join Date: Aug 2014
Location: Featherstone, West Yorkshire, UK.
Posts: 386
Default Re: EQ circuit testing

Yes, strange waveform.

If you overload an input, you should get a clipped sinusoid, not quite a square wave, but close. In your case, you still have some semblance of sinusoid on the positive excursion, but the negative is missing completely.

If you are inclined to investigate further; deliberately overload the input and follow it through to find the stage where the non symetrical clipping starts. To me, it suggests a transistor stage poorly biased, so that it can amplify positive signals, but clips on negative.

Kevin
kevinaston1 is offline  
Old 21st Oct 2017, 8:20 am   #6
Eichberger
Triode
 
Join Date: Apr 2017
Location: Winchester, Hampshire, UK.
Posts: 42
Default Re: EQ circuit testing

This is really interesting. Will do the testing you suggest later. I have also noticed the scope trace flicks up and then settles back to zero when there is no signal. Was thinking there might be a DC voltage on the output.
Eichberger is offline  
Old 21st Oct 2017, 8:50 am   #7
Eichberger
Triode
 
Join Date: Apr 2017
Location: Winchester, Hampshire, UK.
Posts: 42
Default Re: EQ circuit testing

Just confirmed that I've got almost 14v DC across the EQ output. Could this and the non-symmetrical distortion be part of the same problem?
Eichberger is offline  
Old 21st Oct 2017, 10:31 am   #8
Eichberger
Triode
 
Join Date: Apr 2017
Location: Winchester, Hampshire, UK.
Posts: 42
Default Re: EQ circuit testing

So I've done some more testing. I have attached the EQ circuit diagram and EQ curves if these are useful. The tests were done on channel one but I got similar results in a quick check of channel 2.

At the EQ output there is +13.8v DC bias. When the input signal overloads, the negative cycle begins to clip long before the positive cycle changes. Clipping begins at about 16v peak-to-peak. The positive cycle does not clip as such but does eventually start to distort.

Where the collector of Q106 connects to the base of Q108 there is similar distortion but the signal is much smaller and inverted. There is also a DC bias here of around 2 or 3 volts (Sorry forgot to note how much exactly).

At the input there is no distortion and no DC.

The other thing I noticed is that the EQ circuit is providing around 34dB of gain between input and output at 2kHz. The EQ curves (also attached) show 0dB at this frequency. I also tested at 10 kHz and 16kHz and found that there is the right amount of gain relative to frequency (e.g. an additional 8dB between 2 and 10kHz).

Should the EQ circuit be giving so much gain? There is gain stage before it giving around 40dB of gain. If I'm understanding the circuit correctly the line input is reduced down to microphone level by a 1Megohm series resistor.
Attached Thumbnails
Click image for larger version

Name:	Screen Shot 2017-10-21 at 09.30.16.jpg
Views:	94
Size:	44.3 KB
ID:	151083   Click image for larger version

Name:	Screen Shot 2017-10-21 at 09.53.32.jpg
Views:	67
Size:	74.1 KB
ID:	151084  
Eichberger is offline  
Old 24th Oct 2017, 10:04 pm   #9
ricard
Octode
 
ricard's Avatar
 
Join Date: Mar 2006
Location: Lund, Sweden
Posts: 1,631
Default Re: EQ circuit testing

I'm assuming the 13.8V DC bias is on the emitter of Q109, not at the right hand side of C134.

It would be reasonable to assume that Q109 is biased at the middle of the supply voltage, so with a supply voltage of 27V, 13.8V would be about right. Doing some calculations, this means that the emitter current is 13.8mA (13.8V/R144=1kohm), which also makes sense considering the base resistor R141 is 220k and the BC107B has a gain of 200-400, so the ratio between the currents flowing through R144 and R141 will be the same as the gain as there are no other sources of bias for the transistor. With the emitter (and hence the base) being around mid-VCC, this means the ratio between the resistor values will be roughly the same as the gain which it is. This transistor is not providing any gain in itself, it's an emitter follower just acting as a buffer.

It's hard to say at a glance what the gain of the EQ amplifier is, but certainly there is a darlington pair at the input, so there will certainly be a lot of raw gain to start with, but then of course there is a lot of feedback too.

Regarding the input, since there is really only one input on the amplifier, it's not uncommon to lower the signal level for line level signals using a simple series resistor or divider, rather than skipping the first amplifier stage for high level signals.

What serial number does the machine have? There is an inherent problem with the circuitry on machines with serial numbers below 2209550 which causes erratic connections in the record/playback switches, usually manifesting themselves as loud scratches and pops in the audio channel when the switches are moved between their three positions. I'm thinking that the problem with clipping could be due to an erratic connection causing too much gain in the EQ amplifier.

Last edited by ricard; 24th Oct 2017 at 10:15 pm. Reason: Added note about sub-serial-2209550 problems.
ricard is offline  
Old 25th Oct 2017, 11:48 am   #10
Eichberger
Triode
 
Join Date: Apr 2017
Location: Winchester, Hampshire, UK.
Posts: 42
Default Re: EQ circuit testing

Thanks for jumping in ricard. A forum member has very generously been helping me with this and we've established that the EQ circuit is working well. It's looking like the problem is most likely with the head which is a shame as I really like the machine.

Mine is the older version with the problematic switches but I've cleaned the contacts and verified they are working ok

The last roll of the die is to change the Tantalum caps that connect the heads to the playback amp in case they are leaky. Then it might be time to get on eBay and find something else!
Eichberger is offline  
Old 25th Oct 2017, 8:52 pm   #11
ricard
Octode
 
ricard's Avatar
 
Join Date: Mar 2006
Location: Lund, Sweden
Posts: 1,631
Default Re: EQ circuit testing

In what way does it seem to be the head? Open circuit? That must be rare, I've very rarely seen a Tandberg head go open circuit (the only time I've encountered one was one half of the erase head on a model 5). Do both halves seem to be the same?
ricard is offline  
Old 26th Oct 2017, 2:59 am   #12
Eichberger
Triode
 
Join Date: Apr 2017
Location: Winchester, Hampshire, UK.
Posts: 42
Default Re: EQ circuit testing

Nothing like that. I think the gap might have started opening up as I am getting loss of high frequencies. 10db loss between 5kHz and 10kHz then very little HF beyond about 14kHz
Eichberger is offline  
Old 26th Oct 2017, 7:39 am   #13
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: EQ circuit testing

The circuit design around that input darlington isn't very good. Q106 gets its DC bias from a feedback path via R123.

So, either the collector-base voltage of the darlington is very low (which increases distortion and early effect, as well as severely limiting the swing) or else the circuit is relying on DC voltage drop in R123 to allow the emitter of Q108 and hence the collectors of
Q106/Q107 to lift a bit. If so, this depends on the base current of Q106. This depends on the product of the current gains of Q106 and Q 107 which will have at least an order of magnitude of uncertainty. They should have known this back in the day. What they may not have known would be the limited life expectancy of very high value resistors.

Check R123

Any leakage in C119 would cause problems, too.

As has already been said, the waveform posted suggests a lot too much gain or a lot too much input signal. Lots of amplifier circuits behave badly coming out of clipping. Some recover surprisingly slowly because their bias circuits have been disturbed by currents while clipped. Some hang about and come out with a snap, some go unstable and give a burst of oscillation.

Music signals have quite high peaks, plus there are 'fridge clicks and other transients, so the sane end of the audio designer spectrum learned to design circuitry which can handle occasional clipping and comes out of clipping cleanly and does not extend the problem.

So don't be much surprised at odd behaviour if an older transistor circuit gets driven into clipping. There has been some real progress made in circuit design, but it's been quiet and lost in all the smoke and thundering of phoolish egomaniacs.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 27th Oct 2017, 9:27 am   #14
Eichberger
Triode
 
Join Date: Apr 2017
Location: Winchester, Hampshire, UK.
Posts: 42
Default Re: EQ circuit testing

Thanks David. I'm fairly sure the EQ circuit is working as intended and the high frequency losses are occurring at the play/record head.

The gap is not visibly enlarged and I did have it relapped by this place: http://www.summertone.com/services/

They said it was fine for relapping though so I'm still a bit unsure about the whole thing!
Eichberger is offline  
Old 27th Oct 2017, 2:05 pm   #15
TIMTAPE
Octode
 
Join Date: Jul 2006
Location: Perth, Western Australia
Posts: 1,965
Default Re: EQ circuit testing

They should have done a good job. It's also very important to ensure the tape is contacting the head gap optimally. If there's even a slight loss of intimate tape-to-head contact, say a cigarette paper's width, a lot of the treble will disappear.

I probably said it before but as per the manual, any frequency response tests must be recorded well below maximum level on the record meter. In practice the meter should be barely registering the signal, which corresponds to about 20db below nominal maximum. The reason is as you should have worked out from the record EQ circuit: the recorded high frequencies are already greatly boosted above the mids and lows and will reach their limit well before the mids and lows do. But normally that isnt apparent on the record meter, although interestingly on some Tandberg models it does, which is unusual.
TIMTAPE is offline  
Old 30th Oct 2017, 11:24 am   #16
Eichberger
Triode
 
Join Date: Apr 2017
Location: Winchester, Hampshire, UK.
Posts: 42
Default Re: EQ circuit testing

Thanks TIMTAPE. I'm fairly sure the tape to head contact is good.

When I was testing the recording EQ circuit I did notice that the magic eye meters do show the signal level post EQ so less chance of overloading the high frequencies.
Eichberger is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 8:15 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.