switching balanced & unbalanced output on 388

mr.blisset

New member
I've always run the 388 master outs unbalanced as I can see in the schematics theres bunch of things going on to convert the master out to a balanced signal. I've always thought switching to balanced will degrade the audio path quality. However, I wouldn't mind strapping a comp or eq across the mix. That gear is balanced of course.

I have a Radial Pro DI 2 which is a passive DI box, would this work better to make the conversion? Or for better quality would I be better off using a couple of Lundahl Problem Solvers? Maybe getting some good transformers and making my own box?

thb Im not hearing a big difference when switching between balanced and unbalanced. Are the any test I could do that might reveal the differences?

am I completely over thinking this?:rolleyes:
 
I've always run the 388 master outs unbalanced as I can see in the schematics theres bunch of things going on to convert the master out to a balanced signal. I've always thought switching to balanced will degrade the audio path quality. However, I wouldn't mind strapping a comp or eq across the mix. That gear is balanced of course.

I have a Radial Pro DI 2 which is a passive DI box, would this work better to make the conversion? Or for better quality would I be better off using a couple of Lundahl Problem Solvers? Maybe getting some good transformers and making my own box?

thb Im not hearing a big difference when switching between balanced and unbalanced. Are the any test I could do that might reveal the differences?

am I completely over thinking this?:rolleyes:

Well "I" can't see from any schematics! I cannot see there could be much 'adulteration'? If an op amp there will just be another stage. If discrete a few more transistors or/and a transformer.

I personally think Lundies or even Jensen traffs are a bit of an overkill for tape signals. The distortion levels of £20/$20 traffs from the likes of OEP are well below those of tape, where do you peak to? 3% thd? Bandwidth of '600-600Ohm' output traffs is not a problem because they are 1:1 and often bifiliar wound.

In the esoteric would of digital recording a bog standard balanced INPUT is some 14dB noisier than an unbalanced one but since even the former come out below -100dBu not really a problem?

A DI box is 10:1, not at all what you need.

Dave.
 
The 'bunch of things' is very unlikely to introduce any problems, and I doubt if you could even hear it - probably a chip with differential output. If you are going into an analogue external device, that's degrading the signal far more than the balancing would, and if you select unbalanced out then you're not taking advantage of the signal to noise the desk has, and any gain makeup will be done at the processor, where it will have to make the gain up anyway. I've never heard anyone concerned about their mixer sounding worse when balanced - probably because they don't. The unbalanced external linkages would probably raise the noise floor more than if you run balanced, that's why people use it through choice.
 
The balance amp on the 388 is a +/-15V powered semi-discrete design with a discrete transistor output stage, and the input stage uses a 2041 opamp, which is considered a high output variant of the TL072. In my opinion I’d leave it alone and just use what’s in the 388. I’m not sure what you would gain by adding boxes in between the 388 and your downstream device.

You might have said it already but what are you driving withbthe balanced out specifically? What device are the 388 outputs connected to?
 
You're not going to outthink the Tascam engineers. You've got a balanced output and a balanced input. Connect them, they were literally made for each other.
 
Err? They SAY "sarcasm is the lowest form of wit" and it seems to have done me no good!

To be plain I would LOOOOOVE to see those schematics!

Dave.
 
The balance amp on the 388 is a +/-15V powered semi-discrete design with a discrete transistor output stage, and the input stage uses a 2041 opamp, which is considered a high output variant of the TL072. In my opinion I’d leave it alone and just use what’s in the 388. I’m not sure what you would gain by adding boxes in between the 388 and your downstream device.

You might have said it already but what are you driving withbthe balanced out specifically? What device are the 388 outputs connected to?

According to this: http://52ebad10ee97eea25d5e-d7d4081...48.r84.cf3.rackcdn.com/UK_NJR_NJM2041D_DS.pdf

The IC is a bipolar input device but seems to have the poor load drive capability of the '72 (hence the extra transistors). It seems an NE5532, better an LM4562 could be a drop in, the discrete kit scrapped and still get all the drive you need into 600 Ohms? NOT that anybody does that these days!

But as you say Mr S, why bother?

Dave.
 
I mean, you're not going to hurt anything running the unbalanced output into a balanced input either, but balanced almost has to be better for noise. Not that you'll notice under the waterfall of tape hiss...

A DI (as mentioned above) will attenuate the signal and add its own little bit of noise. You'll pick up less (bot not no) noise on the balanced run, but you'll have to turn that noise up more at the other end.
 
I mean, you're not going to hurt anything running the unbalanced output into a balanced input either, but balanced almost has to be better for noise. Not that you'll notice under the waterfall of tape hiss...

A DI (as mentioned above) will attenuate the signal and add its own little bit of noise. You'll pick up less (bot not no) noise on the balanced run, but you'll have to turn that noise up more at the other end.

Try as I might Ash' I cannot find a number for the ratio of that DI nor an input Z. It claims "600 Ohms" outputs but how can they say that unless ratio and source Z are specified?

If we are generous and say keyboards can have a 1kR source Z that means at best a 1.3:1 reduction and thus nearly 3dB signal loss. V poor information IMHO for such expensive and well thought of gear.

Dave.
 
The specs on their website say 140K input and 150 Ohm out. That's still meaningless, of course. It can't possibly be unity through, though. It's never a good idea to turn something down just to turn it back up again at the very next stage.
 
The only reason to use the balanced output is if there are issues with RF noise interference in the cable run, or if the balanced out achieves better level matching with the device connected to the outputs. If the nominal level of the load is -10dBV, and the cable runs are short and there's no problem with RF interference, then the best set of outputs are the unbalanced outs. If the inputs on the load device are low impedance and 0dBu or greater nominal level then the balanced outs would probably be a better option...if the inputs on the load device are balanced only then the balanced outs are the only option. But the cleanest outputs from the 388 main buss are the unbalanced -10dBV outs...that is the native operating level of the 388 and that path involves the least amount of circuitry. But the 388 has the balanced out option and there's nothing wrong with it...like bouldersoundguy said, Tascam included it in the feature set for the exact reason you are wanting it for, and there's nothing wrong with the amp design. Yes you could spend time and money and introduce some external unit along with the potential for grounding issues, hum, noise, level mismatching, etc., or just use what's already there, keep it simple. The choice is yours.
 
If you really want good sound quality, why are you using cassette?

Not cassette...1/4" 8-track open reel...yes...that's about the same track width as 4-track cassette, but its running at 7.5ips...its enough track "real estate" to qualify appropriate dbx n/r system as Type I vs Type II. :D
 
Not cassette...1/4" 8-track open reel...yes...that's about the same track width as 4-track cassette, but its running at 7.5ips...its enough track "real estate" to qualify appropriate dbx n/r system as Type I vs Type II. :D

I stand corrected. And I'd point out that Tascam is even less likely to screw up the balanced outputs on that machine than on a cassette based machine.
 
Yeah...They used that semi-discrete circuit in various iterations on a bunch of different units starting back in the early 1980s...The 52, 58, MS16, LA-40, M-512/M-520 for instance...it was a well-established design by the time the 388 hit the market.
 
The 388 is not a cassette deck but it might as well be if the tracks are that narrow. In working with such a compromised system, it make no sense whatsoever to go nit picking at the circuits used or what is in the unit. If you were after better quality the first place you should look is at a 1/2" 8 track format rather than pushing the limit of technology with a proprietary head you will never be able to find to replace the worn one you have.
It reminds me of a person with a Plastic Porta Studio trying to do the work that belongs on a 2" machine. You are never going to get the results of the better equipment at this lower cost level.
 
The 388 is not a cassette deck but it might as well be if the tracks are that narrow. In working with such a compromised system, it make no sense whatsoever to go nit picking at the circuits used or what is in the unit. If you were after better quality the first place you should look is at a 1/2" 8 track format rather than pushing the limit of technology with a proprietary head you will never be able to find to replace the worn one you have.
It reminds me of a person with a Plastic Porta Studio trying to do the work that belongs on a 2" machine. You are never going to get the results of the better equipment at this lower cost level.

Well, not quite. The 4X speed increase gives a near 6dB noise advantage then wow and flutter is easier to control and (I assume?) no pressure pads will drastically reduce scrape flutter sidebands.

If a 3 head machine the rec and play heads can be optimized for the job. The higher speed means the HF cut off point is higher for a wider replay gap giving another noise advantage.

In any case, at the end of its development the cassette machine was remarkably good. People like Nakamichi extracted fabulous performance when using top grade tape. I have had a Denon (auto tune) that gave very close to CD quality with TDK SA and now have a Sony S machine (that I need to fit belts to!) that is pretty close.
IMHO cassette at its best beat vinyl! (Ooo! In "it" now!)

Dave.
 
Back
Top