The Bias setting?

[RichTassoni]

You have a deck that you can tweak the bias and level settings such as my TEAC V5000 or just the bias, the manual says before recording set the bias setting to the center position because the deck knows the type of tape you're using and automatically sets the bias, I'm using TDK SA,SAX, MAXELL UD XLII HIGH BIAS. Another part in the manual tells you how to set the level tone first than the bias using the metters to that specific type of tape. So do most of you tape heads tweek your bias or leave the settings in the center and let the deck just automatically set the bias by the type of tape you're using, I tried it both ways can't say I hear any difference. Thanks for anyone's advice on this subject.

 

[oldvinyldude]

I do not own a deck with an autobias function, but a few I own do have an BIAS finetune pot. Evidence\experience suggests that a large percentage of decks made from the early 80's onward were adjusted at the factory for XLII tape-when the CRO2 setting was engaged (or later decks and tape shells provided an auto setting for this). Most other CRO2 tapes had very similar BIAS requirements as well, so most sounded very good.

Most of my decks are 3 head, so when recording, I do tweak the available BIAS finetune to find my preference. This adjustment seems to have less effect when the deck is set to use CRO2 tapes. When using type 1 tapes, the effect is quite pronounced, and I am often adjusting for less BIAS than the center detent.

 

[RichTassoni]

I do not own a deck with an autobias function, but a few I own do have an BIAS finetune pot. Evidence\experience suggests that a large percentage of decks made from the early 80's onward were adjusted at the factory for XLII tape-when the CRO2 setting was engaged (or later decks and tape shells provided an auto setting for this). Most other CRO2 tapes had very similar BIAS requirements as well, so most sounded very good.

Most of my decks are 3 head, so when recording, I do tweak the available BIAS finetune to find my preference. This adjustment seems to have less effect when the deck is set to use CRO2 tapes. When using type 1 tapes, the effect is quite pronounced, and I am often adjusting for less BIAS than the center detent.

Thanks for the layman's terms on bias setting I have 4 decks all 3 head the 3 I use most are the TEAC V5000 I'm recording with and play back, 2 ONKYO TA R500 that really sounds great just playing the other a JVC TD V621 that gives great recording and playback.One manual explains it well that adjusting the bias to the - increases the highs in treble and to the right + side decreases the highs lowering the treble highs. Does this sound right to you thanks.

 

[oldvinyldude]

adjusting the bias to the - increases the highs in treble and to the right + side decreases the highs lowering the treble highs. Does this sound right to you thanks.

Yes-this is the most audible of the effects of changing BIAS. Overbias (+) will suppress the high end and underbias (-) will accentuate the high end. The goal is generally to match the high end response between the source and the tape. Some users will underbias slightly to accentuate the highs or overbias to 'warm up\fatten the sound" a bit. There is a theoretically correct setting, and there is a subjective preferred setting-so go with what sounds good to your ears.

 

[RichTassoni]

Yes-this is the most audible of the effects of changing BIAS. Overbias (+) will suppress the high end and underbias (-) will accentuate the high end. The goal is generally to match the high end response between the source and the tape. Some users will underbias slightly to accentuate the highs or underbias to 'warm up\fatten the sound" a bit. There is a theoretically correct setting, and there is a subjective preferred setting-so go with what sounds good to your ears.

Now that's the best explanation of how bias setting works that I've got from anyone, and also my thoughts the same as yours as what bias does to the tapes and sound changes, IT'S ALL IN WHAT (OUR) EARS HEAR. Thanks.

 

[skywavebe]

Bias adjustment on deck is usually limited to a few dB of effect and since the human ear detects change at about 3dB it most the time is hardly noticed. I test these when a deck is in the shop so I know that they are a marketing item not for real Technicians who know how to Bias a deck. Plus there should be a left and right one if they were serious.

Now deck that have full range bias are those like the Tascam 122 MK II. It has it own oscillator and automatically increases the gain on the meters so that the -20dB recorded signal is at 0 Vu for you. Bias is most the time done on cassette due to saturation and the slow tape speed at -20dB below 0 VU. If your 0 Vu is specified at -10dBu then the signal is at -30dBu.
The decks like the V900X that have Auto Cal did work but they required the Technician to calibrate the deck internal bias adjustments first accurately and then do the time consuming Auto Cal adjustments after that. It take 10-11 second for each trial recording on a tape which is why it is time consuming. The deck if done accurately can come withing 1dB of what a Technician can do. Why 1dB? it has to do with the steps of the adjustment process- a Technician turns a trimmer with infinite resolution but the computer has to do it by steps in a successive approximation method. The steps there are only 1dB or so. It can only come as close as it's resolution allows. 1dB is pretty good though.
I use a tape that is Metal called Sony XR. It is 2dB hotter in record than other metal and other tapes in general. In some decks where there are only one record level set then this tape will stick out like a sore thumb level wise. Well, I put the deck through auto cal and then used that setting and it corrects for the 2dB extra level to make it 0 dB level again- that is the purpose of having such a thing. Other companies may have system like this but not as exact. I have a number of V900X decks as it was TOTL that year.

 

[RichTassoni]

I posted an update photos of my stereo system with my 2 recently purchased cassette decks 1 a ONKYO TA R500 very nice clean sounding, 2 a TEAC V5000 the TEAC failed a week after receiving it sent it back for full refund. Just purchase and on the way a mint looking ONKYO TA 2600, bought one brand new in 1990 best cassette deck I ever had wished I taken better care of it, hoping this one will give me many years of great recording like the that one did if so it will be my main recording deck.Also have in the rack a JVC TD V621 also a really nice deck I bought about 6 years ago and still alsome recording and sounding deck. Also very nice DENNON 800A that's been running great for me about 4 years I've had it. And I just put a bid on a descent looking ONKYO TA R401 for 80.00 thinking I wouldn't win the bid, well I won and it's on the way also. Il demagnetize the heads give them a good cleaning and see what happens wish me luck. If these decks end up working good il repost profile photos.

 

[Late Boomer]

This is the continuation of a discussion started in Recording level, refresh my memory thread. Moved here, as I was asking about bias, not recording level.

--

@WernerO, Sadly, your image is too low res, anyway. If we concentrate on these two curves, I believe the black one is response at 315 Hz, the red one... is it 10 kHz or 12.5 kHz? Seems like it is 12.5 kHz. They are recorded at -20 dB relative to 250 nWb/m, which is 1.2 dB higher than the Dolby level. (On my Technics deck 0 is 2 dB below Dolby level, or 3.2 dB below 0 dB DIN).

If you go from lower bias to higher, S12.5 goes from being higher than S315, then equal to it (at 0 bias setting of course, as this is a reference tape), then lower than S315. So, at -20 dB the responses from S315 and S12.5 are equal to each other AND they are equal to the input signal, -20 dB. This sort of "double" equality ensures correct Dolby tracking. Thirty or so years ago I read that a deck with fewer than 20 positions is useless for precise metering. Your BX300 has 20 positions, and it has five of them below -20 dB.

On my deck, -20 dB is the lowest indication, so I cannot really run the test at -20 dB. So, my idea is to run this test at higher level, closer to the middle of the meter, where I have four notches above and four notches below the middle position of -3 dB (which is -6.2 dB relative to DIN). Will my test will be invalid in relation to the reference data obtained at -20 dB DIN? Oh, man... Maybe I can do it at -10 dB on my scale (that is, -13.2 dB DIN) and it will be good enough?

Reading about your test setup, you write, "For each tape under test first the bias level is determined by making a frequency sweep at -20dB (relative to 200nWb/m ANSI), tuning the deck's front-mounted bias pot to obtain the flattest response below 10 kHz, while still reaching beyond 20kHz, and with up to 3dB peaking accepted above 10kHz." So, it is -20 dB relative to Dolby, not relative to DIN, for which you showed the reference tape specs. I wonder, why?

I don't do a sweep, I use two tones, one channel 400 Hz, another channel 8 kHz, just like some Sony decks. My entry-level deck is spec'd 40-14k within 3 dB, the Sony is 30-15k one deck, 20-17k another deck, within 3 dB (Type I, no Dolby, nice!). The question is, how it will correspond to the reference curve, and what sort of correction I need to make? Is this a simple linear correction?

Comparing results done at -10, -6, and -3 on my scale, I can see that the "window" with equal levels moves towards higher bias. That is, higher input level requires higher bias. Is this expected behavior?

Now, what is the goal of this fiddling. If there is clear difference in treble vs bass response depending on bias, then one would set it where the output is equal, maximizing the bias. Say, in this case 0 setting is the best.

But if there is no difference for any of, say, 0 through 5 marks in the bias knob, which setting should I choose? Higher bias is supposed to give lower distortions, but I cannot verify this improvement (unless the distortions are really gross), also I've read elsewhere that there may be other sorts of distortions that are not defined by the standard, but people hear them (yeah, yeah, golden ears).

My life back then was simpler as my deck did not have bias adjustment Heck, when I was a schoolkid I did not even have a proper deck.

I guess, since I cannot test at exactly the same conditions as the DIN reference, I should rely on my ears after all. But to me the point of this evaluation is to obtain a proper setting without doing multiple takes and comparing them. My deck is only a two-header, so I cannot do it in realtime.

 

[WernerO]

This is the continuation of a discussion started in Recording level, refresh my memory thread. Moved here, as I was asking about bias, not recording level.

But this is not your thread, so why move it here?

Sadly, your image is too low res, anyway.

The image was only there to illustrate the many parameters that depend on absolute bias level, often in a conflicting way. Even so I managed to find a more legible one, this time TDK SA (year unknown).

If we concentrate on these two curves, I believe the black one is response at 315 Hz, the red one... is it 10 kHz or 12.5 kHz? Seems like it is 12.5 kHz. They are recorded at -20 dB relative to 250 nWb/m, which is 1.2 dB higher than the Dolby level.

The difference with Dolby level is immaterial. These curves will essentially remain the same, whether referred to 250nWb/m or 218nWb/m.

If you go from lower bias to higher, S12.5 goes from being higher than S315, then equal to it (at 0 bias setting of course, as this is a reference tape), then lower than S315. So, at -20 dB the responses from S315 and S12.5 are equal to each other

Yes. Although I would be careful with 'of course'. Before the application of record equalisation (which is a property of the deck's design) the S315 curve generally does not intersect S10 and higher. You can see this in the TDK SA graph below. Optimal bias is determined with other criteria than frequency response.

Bias only becomes a tool for flattening of the frequency response in a complete deck, in the hands of a consumer.

AND they are equal to the input signal, -20 dB.

Yes, but not by virtue of this particular bias setting. Outside the scope of this graph the deck has been engineered to have input and off-tape signals equal at this setting for this tape.

This sort of "double" equality ensures correct Dolby tracking.

You seem to believe that getting bias set correctly somehow also ensures correct level setting. It does not.

Thirty or so years ago I read that a deck with fewer than 20 positions is useless for precise metering. Your BX300 has 20 positions, and it has five of them below -20 dB.

The BX-300 has 10 positions. They are crude and they are inaccurate.

On my deck, -20 dB is the lowest indication, so I cannot really run the test at -20 dB.

The deck's meters should never be relied on for this sort of measurement or alignment. Possible exceptions are needle instruments or bargraph meters with a zillion segments (e.g. Nak 1000ZXL), when used in a calibration mode that maps the entire meter range onto a few dB. No, measurements like these have to be done with external instruments of much higher resolution, in my case a computer running audioTester.

position of -3 dB (which is -6.2 dB relative to DIN). Will my test will be invalid in relation to the reference data obtained at -20 dB DIN? Oh, man... Maybe I can do it at -10 dB on my scale (that is, -13.2 dB DIN) and it will be good enough?

Invalid. There is a reason this is done at -20dB, the reason being treble saturation setting in at higher levels. You don't want to battle saturation with reduced bias.

Reading about your test setup, you write, "For each tape under test first the bias level is determined by making a frequency sweep at -20dB (relative to 200nWb/m ANSI), tuning the deck's front-mounted bias pot to obtain the flattest response below 10 kHz, while still reaching beyond 20kHz, and with up to 3dB peaking accepted above 10kHz." So, it is -20 dB relative to Dolby, not relative to DIN, for which you showed the reference tape specs. I wonder, why?

Because I standardised on Dolby level and not on 250nWb/m when I started this work. And I did so because at that time all of my decks had meter-zero at Dolby level, and because I had Dolby level tapes and not 250nWb/m tapes. At -20dB a difference of 1.2dB up or down is entirely negligible anyway.

I don't do a sweep, I use two tones, one channel 400 Hz, another channel 8 kHz,

How do you know if 8kHz is a wise choice for a particular deck without running a full sweep?

how it will correspond to the reference curve, and what sort of correction I need to make? Is this a simple linear correction?

What reference curve? What correction?

Comparing results done at -10, -6, and -3 on my scale, I can see that the "window" with equal levels moves towards higher bias. That is, higher input level requires higher bias. Is this expected behavior?

Sorry, but you are not making any sense here. You are 1) over-analysing and 2) not using appropriate measurement methods. The result can only be confusion.

Now, what is the goal of this fiddling. If there is clear difference in treble vs bass response depending on bias, then one would set it where the output is equal, maximizing the bias. Say, in this case 0 setting is the best.

How can you 'maximise the bias' and at the same time claim that '0 setting is the best'???

But if there is no difference for any of, say, 0 through 5 marks in the bias knob, which setting should I choose?

See attached tape curve. S10 and higher always are strong functions of bias current. If you cannot see any difference then this only means that your measurement method is inadequate.

Higher bias is supposed to give lower distortions, but I cannot verify this improvement (unless the distortions are really gross), also I've read elsewhere that there may be other sorts of distortions that are not defined by the standard, but people hear them (yeah, yeah, golden ears).

The standard does not 'define any distortions'. It just lists a particular type of distortion(*) that is a function of bias and that is easy to measure. And this had nothing to do with audio mythology.

(* Actually the dominant distortion factor in magnetic recording: third harmonic.)

I guess, since I cannot test at exactly the same conditions as the DIN reference

You do not have to obey DIN (actually 'IEC', 'DIN' is a misnomer) conditions, but you have to be close to them because you have to remain out of the saturation region. With what you are attempting to do you are way off target, and blind to boot.

Best solution in your case would be to set bias by ear, recording FM interstation noise, pink noise, or even white noise, at not too high a level, audibly comparing source and tape.

To summarise
1) on a deck with only external bias control all you can (have to!) use this knob for is to flatten the off-tape frequency response. Distortion level and record level (=sensitivity) follow automatically (the relations are carved in stone by the deck's design) and will not necessarily be optimal for the particular tape in use.
2) approaching this subject matter objectively requires the correct tools, operated at the correct absolute level
3) doing more than the above is possible, but only with decks that are equipped with the proper trimmer (bias, record level, record equalisation), or when designing or redesigning a deck. None of these seem to apply in your case.

 

[Late Boomer]

This thread because I don't like starting new threads if one for the same topic exists.

Yes, I am over-analytical, but I don't want to be overly technical, so I want to use the level meter. I want to behave as a well-informed consumer

I am maximizing the bias on per tape basis, if you look at the graph it is the position where high and low test signals have the same output level. It is no different than using white noise, but in case of just two frequencies I can plot a pretty graph and find an intersection. Also, it is much easier to do on a two-head deck. In my testing, for some tapes there may be several intersections with different output levels. Clearly, I want the same level as the input. So, no, record level does not automatically fall into place.

If I were using white noise, my main question would remain: at what level should white noise be recorded? -20 dB? -10 dB? Close to the level I am planning to record the tape?

This is the method used by many a recording engineer:

It is not directly translatable to cassettes, still the gist is the same set 1 kHz at 0, then bring 10 kHz to 0 by using bias and treble level control (which consumer decks do not have). He recommends recording at -10 VU for audio cassettes.

I want to use 8 kHz instead of 10 kHz because even at a higher level this frequency is still well within response range for all decks and tapes. Also, because Sony used it but I don't know at which level. Looking at this picture, it should not matter TOO MUCH which recording level to use until the response starts to compress (I re-numbered your scale to put 0 dB at the Dolby level). Even at Dolby level the difference between 400 Hz and 8 kHz is only 1.5 dB, and the level meter on my deck had 3 dB precision at best, 5 dB at the end of the scale, so should be good enough for a guesstimate.

BTW, could you do a short writeup on how to use audiotester sweep on a two-head deck? I tried and failed.

 

[WernerO]

I want to behave as a well-informed consumer

Then be informed that it is pointless to rely on most deck's meters.

I am maximizing the bias on per tape basis,

You are not maximising the bias. You are setting bias so that some other parameter is maximised or optimised.

If I were using white noise, my main question would remain: at what level should white noise be recorded? -20 dB? -10 dB? Close to the level I am planning to record the tape?

You can answer that yourself by recording at several levels and seeing where treble saturation sets in.

As an aside: white noise brings the risk that you will set standing bias a bit too low because the noise signal is already self-biasing.

I want to use 8 kHz instead of 10 kHz because even at a higher level this frequency is still well within response range for all decks and tapes.

But 8kHz is less under control from the front-panel bias knob than a higher frequency. Moreover you would be essentially blind at higher frequencies. Strange things might happen there, depending on the sort of tape.


As for the level at which to test: the curves you took from my site were made with some of the very best heads in industry. 99% of all decks out there will not reach such performance.


A writeup for using AT with two-headers. Maybe one day, when I have the time ...

 

[Late Boomer]

Then be informed that it is pointless to rely on most deck's meters.

The meters are there for a reason, so I am supposed to trust them to some degree when I am making a recording. I do not want to turn my second-class deck into a measurement station. I want to use a consumer-grade device in a consumer-type scenario. Obviously, my measurements won't be as precise as yours, but I am not intending to. At best, I can use this as a guidance to ear-test two or three bias settings instead of a dozen.

You are not maximising the bias. You are setting bias so that some other parameter is maximised or optimised.

If higher bias (within the limits that the typical bias knob allows) lead to lower distortion, then I do want to maximize bias. It is a multi-variable optimization, where I want to:
* equalize low and high frequency response
* equalize output level to input level
* use the highest bias if the two above are satisfied

Using this picture again, treble and bass are equalized at bias knob setting of -2, -1 and 0, so I am going to use 0. For some tapes I have several sections where bass and treble are equal, but at different output levels.

You can answer that yourself by recording at several levels and seeing where treble saturation sets in.

No direct answer Ok, thanks.

* the curves you took from my site were made with some of the very best heads in industry. 99% of all decks out there will not reach such performance.
* white noise brings the risk that you will set standing bias a bit too low because the noise signal is already self-biasing.

If Naks were the only decently-performing decks, cassettes would not get as big as they did. Your 99% is pulled out of a thin air. But my deck is clearly a second rate compared to the BX300, which is why I am using 8 kHz instead of 10 or 12 or white noise to ensure that I am still in the response range.

But 8kHz is less under control from the front-panel bias knob than a higher frequency. Moreover you would be essentially blind at higher frequencies. Strange things might happen there, depending on the sort of tape.

Sony used 8 kHz, I bet they had a reason. If the tape is weird, then setting the right end using a higher frequency would cause the mids to behave weirdly, would not it? It is better to flatten the most usable region and to let the tape do its thing on the edges. Decent tape will have some reasonable expected fall-off. Crappy tape, well, will be all over the place anyway.

 

[WernerO]

I want to use a consumer-grade device in a consumer-type scenario.

Nice to learn that your obsessing over bias is a standard consumer scenario.

Obviously, my measurements won't be as precise as yours, but I am not intending to.

Intentionally crappy measurements. A thing to be applauded.

It is a multi-variable optimization, where I want to:
* equalize low and high frequency response
* equalize output level to input level
* use the highest bias if the two above are satisfied

Juggling these multiple variables is a job done during the deck's design. Afterwards the fine bias knob only controls one of them.

Using this picture again, treble and bass are equalized at bias knob setting of -2, -1 and 0, so I am going to use 0.

'Equalised' only because your meters are not telling you the truth.

No direct answer Ok, thanks.

Actually the only answer that makes sense.

Your 99% is pulled out of a thin air.

OK, it ends here. If you don't really want to learn then you are also no longer going to waste my time.

 

[Late Boomer]

I am not obsessing. My deck has a knob that should be useable and useful. As a user, I am not supposed to measure voltage on the output jacks. If this were a car, the regular instruments should be enough for driving, I should not connect OBDII and fiddle with internal data to correctly apply the brakes.

The knob controls bias. I am choosing how to optimize the three parameters.

I am surprised you got offended, it was a lighthearted comment. Well, maybe your particular repaired well-serviced deck is better than 99% of all the other surviving decks, I am not going to argue. In any case, it was not important for this discussion. The topic has been exhausted, I found more than I was hoping for. I am grateful for your help and guidance.

 

[twiiii]

All this proves is the short comings of the cassette tape medium. Yes I had multiple Nakamichi to make tapes for others and for the cars that only had cassette machines. Other wise I would have saved my money. My HS B-77's with Deci-linear noise reduction were much better and my A-77 IV with Dolby was much better, too. Yes I admit the Dragon was a great machine as were the advanced 700 and 1000 models. But even at $3600.00 list price a nakamichi could never match my B-77's. Let alone my professional CD recorders and professional Sony Beta machines recording audio. If you want to stick with tape try a Sony Beta Hifi machine model SL 1800. JVC's 9000 series S-Vhs machines do a great job, too for recording audio. I used my Nakamichi 550 that was portable with Audio Technika condenser mics to record live steam trains. I tried to keep the machine from going into limiting which was a nice feature you could turn on or off. All the tapes have been dubbed to CD.

 

[Late Boomer]

Here is a video for my very unscientific experiment

And a screenshot:

 

[WernerO]

No harm done.

Nice video, particularly your pointing out that chromiumdioxide is long-term instable, instead of reverting to the myth that chrome is something special for connoisseurs.

But no, TDK SA in the 1970s was not made for half-way compatibility with type I. It is simply so that absolute bias levels, regardless of tape type, have crept upwards through the years.

 

[Wilhem]

Now that's the best explanation of how bias setting works that I've got from anyone, and also my thoughts the same as yours as what bias does to the tapes and sound changes, IT'S ALL IN WHAT (OUR) EARS HEAR. Thanks.

The following document is a more detailed explanation of what bias is and does, written in layman's terms.

http://www.ant-audio.co.uk/Tape_Recording/Library/Bias.pdf

 

[Joseph G]

Hello, I just wanted to throw this out there. When I optimized bias on my 3 head cassette decks, I would always use FM noise from a tuner. But before any adjustments are done, one should properly demagnetize and clean the heads. On a cassette deck, the record and play gaps are very tiny, and the slightest hint of oxide will have a huge impact on sound quality. So before using a sine wave generator, use the FM noise, and check it at -10dB. For metal tape, you can go higher, like -3dB. This gets you in the ballpark much faster than using the signal generator alone. Also, for best distortion specs, the sound will be not as bright as the input. It’s all a compromise!