Sanity Check - Convert 1/2" 4 Track to 2 Track (With a twist)

[bjoneson]

Ok.. I have a 1/2" 4  Track Scully 280B deck sitting in my garage. Transport is running like a top, and have nice SAKI glass / ferrite heads that are seemingly indestructible.

I have no use for a 4 track deck. But do want a decent 2 track deck for mixdown. I already have another 280B in 1/4" I use for this regularly.

Nortronics (Scullys OEM head supplier), never made 1/2" two track heads for this machine. Custom manufactured heads are out of the price range I can justify for this project.

My idea is to build custom electronics to drive tracks 1/2, and 3/4 in parallel (or series, not sure which makes the most sense yet). Basically a "poor man's" 1/2" deck, but with true stereo electronics.

Everything would work and behave as a typical 1/2" two track, except there would be a guard band in the middle of each track.

If my goal was to have a fantastic 1/2" stereo deck, I would go buy a studer. That's not a feasible investment for me.

My goal is to make the best use of the stuff I have laying around, and keep myself occupied through the winter months.

I just need a sanity check on whether what I'm proposing is bordering on insanity.

I'm comfortable enough with theory and PCB fabrication to get through the design aspect, though I'm sure questions will arise. Has anyone done anything like this before? Am I out of my mind?

 

[ruffrecords]

Designing record/replay electronics is a non trivial exercise. I think you need to ask yourself in what way would this be better than driving two channels with the same input and combining the outputs?

Cheers

ian

 

[bjoneson]

My theory is that it would minimize phase or frequency response anomalies from the 2 independent record and repro amplifiers.

My intent isn't to design the electronics "from the ground up", just make adjustments for the change in impedance associated with using multiple head windings.

 

[ruffrecords]

My theory is that it would minimize phase or frequency response anomalies from the 2 independent record and repro amplifiers.

My intent isn't to design the electronics "from the ground up", just make adjustments for the change in impedance associated with using multiple head windings.

OK, understood. I was going to say that I expected the differences in the electronics to be less that between two heads in a stack but that does not matter because if it is true then one electronics path or two will not make a difference. It might be worth recording some tones from a single source onto two tracks and looking at the phase between them on replay just to get an idea of how much phase shift there is.

Cheers

Ian

 

[gridcurrent]

I just need a sanity check on whether what I'm proposing is bordering on insanity.

Paralleling tracks for reduced noise was discussed over at GS.
There were arguments.
Ask Jay McKnight.

 

[bjoneson]

Yeah, I actually participated in that thread briefly back in September. Much name dropping, and pasionate bantor but nothing conclusive.

Maybe this whole thing turns in to an experiment, and I can test / measure.

 

[abbey road d enfer]

My idea is to build custom electronics to drive tracks 1/2, and 3/4 in parallel (or series, not sure which makes the most sense yet). Basically a "poor man's" 1/2" deck, but with true stereo electronics.

As Ian said, it's far from trivial. You may want to mod the existing electronics, but you'll be faced with the problem of impedances being either halved or doubled.  Connecting the record heads in series increases the impedance, which would need doubling the record and bias voltage; that would probably be a major undertaking. Parallels connection would require an increase of record and bias current, which may be less difficult to do; in fact I guess the best option wolud be to keep the separate final record stage/bias circuits, and feed them from one single channel. That way you would have only one set of level/EQ trimmers. The replay heads should definitely be put in parallels, because the series connection would make them too sensitive to interference. It may be possible to increase the bias current of the replay input stage for optimizing to the lower impedance.
Just as well, I would keep the separate erase circuits.
Can you post the schemo? then we could elaborate somewhat...

 

[bjoneson]

Full schematic of Record / Repro circuits is attached (it's large, want to make sure you can zoom in as necessary).

The stock heads that shipped with the machine had the following electrical specs:

1/4" Half Track Stereo

Repro (Nortronics 9211):
- Inductance: 100mH (1kHz)
- Impedance: 640 Ohms (1kHz)
- Resistance: 95 Ohms (DC)
- Play Output RMS: 1.3mV (1kHz, 12dB below saturation)
- 10kHz / 1kHz Output Ratio: +2dB

Record (Nortronics 9209):
- Inductance: 5mH (1kHz)
- Impedance: 26 Ohms (1kHz)
- Resistance: 4.2 Ohms (DC)
- Peak Bias Current: 7.5mA (100kHz RMS)
- Peak Bias Voltage: 12V (100kHz RMS)
- Audio Record Current: 600uA (RMS)

Erase (Nortronics 9223):
- Inductance: 0.2mH (1kHz)
- Impedance: 140 Ohms (100kHz)
- Resistance: 3.7 Ohms (DC)
- Erase Current RMS: 115mA
- Erase Voltage RMS: 16V (100kHz)
- DC Erase Current: 200mA
- Erase from Sat.: 70dB (400Hz)



1/2" 4 Track

Repro (Nortronics 9476):
- Inductance: 100mH (1kHz)
- Impedance: 650 Ohms (1kHz)
- Resistance: 150 Ohms (DC)
- Play Output RMS: 1.3mV (1kHz, 12dB below saturation)
- 10kHz / 1kHz Output Ratio: +7dB

Record (Nortronics 9469):
- Inductance: 4mH (1kHz)
- Impedance: 30 Ohms (1kHz)
- Resistance: 10 Ohms (DC)
- Peak Bias Current: 7mA (100kHz RMS)
- Peak Bias Voltage: 10V (100kHz RMS)
- Audio Record Current: 430uA (RMS)

Erase (Nortronics 9470):
- Inductance: 0.17mH (1kHz)
- Impedance: 70 Ohms (60kHz)
- Resistance: 3 Ohms (DC)
- Erase Current RMS: 110mA
- Erase Voltage RMS: 8V (60kHz)
- DC Erase Current: 200mA
- Erase from Sat.: 70dB (400Hz)



1/4" Full Track Mono

Repro (Nortronics 9110):
- Inductance: 100mH (1kHz)
- Impedance: 650 Ohms (1kHz)
- Resistance: 100 Ohms (DC)
- Play Output RMS: 2.0mV (1kHz, 12dB below saturation)
- 10kHz / 1kHz Output Ratio: +2dB

Record (Nortronics 9104):
- Inductance: 4mH (1kHz)
- Impedance: 26 Ohms (1kHz)
- Resistance: 3.3 Ohms (DC)
- Peak Bias Current: 9mA (100kHz RMS)
- Peak Bias Voltage: 14V (100kHz RMS)
- Audio Record Current: 650uA (RMS)

Erase (Nortronics 9128):
- Inductance: 0.12mH (1kHz)
- Impedance: 70 Ohms (100kHz)
- Resistance: 0.7 Ohms (DC)
- Erase Current RMS: 200mA
- Erase Voltage RMS: 14V (100kHz)
- DC Erase Current: 350mA
- Erase from Sat.: 70dB (400Hz)



There does not seem to be any changes in circuitry associated with any of the above head stack configurations.

It's worth noting, I have no idea what the electrical properties of the SAKI 1/2" 4 track heads currently installed on the machine are. They were on it when I got the machine, and I can find no literature or specification sheets anywhere.

Thanks for taking a look!

 

[abbey road d enfer]

There does not seem to be any changes in circuitry associated with any of the above head stack configurations.

That is pretty standard. Studer/3M/Ampex recorders had identical electronics whatever the format.

I have no idea what the electrical properties of the SAKI 1/2" 4 track heads currently installed on the machine are. They were on it when I got the machine, and I can find no literature or specification sheets anywhere.

If you can manage to align the machine correctly, that means the characteristics are close enough.

Now, as often, the things that seem at first to be easisest turn out to be quite a challenge.
As predicted, paralleling the erase heads should probably be avoided, because the bias transformer (which also carries the full erase load) may not be able to deliver AND the LDR that is used for switching may not be capable of carrying twice the load.
But the biggest issue is that, in order to feed simultaneously both final record stages, because they are very low impedance (common-base input stage), it is necessary to replace all the components of the REC EQ by halving all resistors and doubling all caps.

 

[JohnRoberts]

At the end of all this effort you will still have guard bands in the tape and shielding in the heads between tracks if they were capable of single track punch-in.

At best not as good as using the proper 2T head.

JR

 

[bjoneson]

100% Agreed. The question is would it be "better" than, standard 1/4" half track. As mentioned before, a 1/2" 2 track stack is beyond the realm of financial sanity for this deck. Was quoted around $2500 for a set of custom heads from JRF. At that point, may as well start searching for another more modern deck.

Just trying to determine if there is any value in this proposition.

Another idea I've thought of is to simply use it like a 1/2" quarter track stereo deck. Quality would be similar to 1/4" half track, but would potentially save on tape costs. (1/2" tape is currently less than 2x 1/4" tape)

 

[JohnRoberts]

For simple audio performance use digital. For tape compression/distortion the 4T will probably work as is.

Have you tried sending the same signal to 2 tracks at a time, then summing the playback? HF may be affected.

JR

 

[Michael Tibes]

Why mot just try to parallel the rec heads, put the pb heads in serial and see what happens? I'd probably watch the bias to check if it gets loaded down. That's a very quick and hands on thought, no warranty for anything ;-) I just don't immediately see that it could brake something?

Michael

 

[abbey road d enfer]

Why mot just try to parallel the rec heads, put the pb heads in serial and see what happens?

I  explained earlier why that wouldn't be right.

I'd probably watch the bias to check if it gets loaded down.

Not only bias; the whole calibration would need to be done, and I can see how some of the adjustments would be out of range. [/quote] That's a very quick and hands on thought, no warranty for anything ;-) [/quote] I can warranty that the performance would be worse than what it is at the moment. Remember the OP wants to improve performance.

I just don't immediately see that it could brake something?

Indeed, nothing would be damaged, but is it reason enough to do it?

 

[bjoneson]

I'm coming to the conclusion that modifying for parallel tracking / pseudo 1/2" stereo is probably not a worthwhile endeavor.

I think my efforts are probably better focused on tackling other issues with the machine that are more likely to bring measurable improvements.

Those are:

- Ouput Transformer (the stock one is egregiously under specd, and can do a drop in replacement from Jensen)
- Buffer for the meter
- Increase maximum signal level of record preamp (can't presently saturate modern tape before this stage clips)

 

[abbey road d enfer]

I'm coming to the conclusion that modifying for parallel tracking / pseudo 1/2" stereo is probably not a worthwhile endeavor.

I tend to agree.

I think my efforts are probably better focused on tackling other issues with the machine that are more likely to bring measurable improvements.

I would agree too.

Those are:

- Ouput Transformer (the stock one is egregiously under specd, and can do a drop in replacement from Jensen)

Is it a 1:1 or 1:2? In the former case, you may do completely without a xfmr and implement balanced-impedance outputs.

- Buffer for the meter

I can't see where the meter connects but very often it was just bridged to the output. One may argue that the level of distortion introduced by the internal rectifiers is negligible compared to tape distortion, but the character of distortion is very different and can be heard ocer tape distortion.

- Increase maximum signal level of record preamp (can't presently saturate modern tape before this stage clips)

That's a point that deserves some investigation, since the record output stage is capable of much more than the nominal 600uA ( I suspect that for the then-contemporary "operating level" 185nWeber/m, multiply by 2.8 for the hottest tapes).
It appears that, due to te particular topology of the record amp, it is necessary to feed significant current into its low-Z input.
I don't know the values of the EQ components (R113, 112, C109...) but I suspect the limit would probably be the 2-transistor stage around Q4 (it's hard to identify the other components) located SW on you pic.
It would be interesting to hav ethe values of the components on the rec EQ cards, and see if it's possible to scale down the impedance in order to drive more energically the record amp.

 

[gridcurrent]

If you want a quieter tape system, buy a set of Dolbies for 100.00.  Don't let anyone tell you they don't sound good.
If you really want tape saturation, use a lower output tape (456/226/250/468, etc).  Modifying the record electronics so you can "cleanly" saturate is questionable, though the idea likely has fans over at GS.  There is yet another forum,  a mutual admiration society, where the topic was beat to death.
Likewise, modifying the line output stage with a fancier or no transformer, so you can hear the tape saturation 'better" seems a bit off.
Just a bit of history, when 456 and 250 were introduced back in 1975, NOBODY complained about reduced tape compression.  If compression is so cool as perceived to today's knob turners, then high output tapes would not have been so universally embraced.
And we still have the continuing debate about re the distortion induced by VU meters on the line output.  Based on the millions of time tested recordings mastered on machines such as the Ampex 300, Scully 280, others, there is not a shred of evidence that meter distortion is objectionable; it certainly is miniscule compared to tape non-linearity.

 

[JohnRoberts]

If you want a quieter tape system, buy a set of Dolbies for 100.00.  Don't let anyone tell you they don't sound good.

;D  Companding NR trades noise floor for transient or dynamic errors. Dolby(? A, C?) is less aggressive than dbx , but they have pros and cons. (I designed and sold a tape NR back in the '80s, of course I thought mine was best, but when you could buy assembled dbx NR for less than my kit cost, I knew it was doomed). 

I repeat if you want good sound quality just go straight to digital.

If you really want tape saturation, use a lower output tape (456/226/250/468, etc).  Modifying the record electronics so you can "cleanly" saturate is questionable, though the idea likely has fans over at GS.  There is yet another forum,  a mutual admiration society, where the topic was beat to death.
Likewise, modifying the line output stage with a fancier or no transformer, so you can hear the tape saturation 'better" seems a bit off.

saturation plug-in?  different flavors of distortion. 

Just a bit of history, when 456 and 250 were introduced back in 1975, NOBODY complained about reduced tape compression.  If compression is so cool as perceived to today's knob turners, then high output tapes would not have been so universally embraced.

for mastering si, for tracking not as much. Tape compression was kind of useful on individual drum tracks, while  tape saturation on the final mix is just muddying things up. Of course they could just hit the high output tape harder for drum tracks (unless the electronics clip).

And we still have the continuing debate about re the distortion induced by VU meters on the line output.  Based on the millions of time tested recordings mastered on machines such as the Ampex 300, Scully 280, others, there is not a shred of evidence that meter distortion is objectionable; it certainly is miniscule compared to tape non-linearity.

Compared to what? Certainly you don't suggest that the distortion from VU meter diodes hanging on a 600 ohm line was somehow desirable?  I can tolerate some nostalgia for  tape or tube overload distortion not meter diode distortion... nah good riddance to it.

JR

 

[abbey road d enfer]

And we still have the continuing debate about re the distortion induced by VU meters on the line output.  Based on the millions of time tested recordings mastered on machines such as the Ampex 300, Scully 280, others, there is not a shred of evidence that meter distortion is objectionable; it certainly is miniscule compared to tape non-linearity.

The issue with VU-meters being strapped directly across a line level is not restricted to tape recorders; in fact they don't affect the recordings, being placed at the line output, the output stage acting as a buffer. They only affect the output signal. The problem has been recognized for a long time. Indeed, when measured with a THD meter, there is not much difference, but when using more sophisticated equipment, such as FFT analyser, it is clear that the rectification byproducts stand out, and the ear confirms that too. This is more discernable with continuous tones than with percussive sounds.
Buffered meters have been used by the most prestigious manufacturers (Studer, Telefunken, Otari) for a number of reasons, but certainly sound quality is one.

 

[bjoneson]

I don't know the values of the EQ components (R113, 112, C109...) but I suspect the limit would probably be the 2-transistor stage around Q4 (it's hard to identify the other components) located SW on you pic.
It would be interesting to hav ethe values of the components on the rec EQ cards, and see if it's possible to scale down the impedance in order to drive more energically the record amp.

For 15ips IEC, values are:

R106: 20K
R113: 4.7K
R112: 4.7K
R114: 220R

C105: N/A
C109: N/A
C107: .015uF
C106: N/A

The clipping occurs in the rec preamp stage around Q4, as you stated. That's the bottleneck. When that stage clips, I'm still a good 2-3dB shy of the 1% 3rd harmonic spec on modern tape (using ATR). Machine was designed for Scotch 206. I have more rolls of sticky Ampex 456 than I know what to do with, and have no desire to try to operate on NOS tape. I just figure if I'm going to push hot signal to this machine as a stereo master recorder, I'd assume have tape be the first thing to saturate, as it's the least objectionable form of distortion in that scenario. I can loop back to digital and get clean any day. I just think this machine deserves to operate at it's best given the changes in tape output over the years, and want to give it a fighting chance.

I didn't realize just how low the input impedance of the main record amplifier was until I took a closer look after you mentioned it. I'm intrigued, because most of the theory I'm comfortable with is with high impedance inputs.

Thanks for taking a look!


You are also correct in regards to the placement of the VU, it's literally strapped right across the secondary of T2 (which is 1:1), right after the line amp.

As an aside... I'm 31 years old. My first studio experience as a teenager was with blackface ADATs. I missed the entire era of analog tape, but have a tremendous fascination and respect for the great many people who pioneered and cultivated the craft of recording on this medium. It appears to be an era in which recording *engineer* really meant something. I have no misconceptions that tape is magic, and I'm not chasing a particular sound. But I do find modern recording tools make me feel very detached from the engineering side of things. It's too easy I think. Knowing you're working with devices and tools that real people, with real hands, who take real pride in their work, spent countless hours creating (as opposed to an assembly line in china) has an impact on the quality of work you do. At least it does for me.