biasing a 388 for 0db tape

M

mr.blisset

New member
My 388 sounds great and it's always been reliable. However, I bought it straight out of a storage locker and I have no idea if the last person to use it was using a different tape with a higher bias. It came with a misc tape on a maxell reel and a box of bad 456 so no clues there. I'm also interested in doing this as I'm hoping to learn something and get better at fixing my own gear in the process.

As I understand it, the manual's method of biasing the 388 is very different from the usual methods when calibrating a three head machine. For instance, from looking at the manual, I've noticed I would only need a MRL test tape if I was attempting to calibrate the reproduction levels and bias. The only checks ive done so far is to record at 0db and note that everything recorded at 0db appears to play back at 0db. It would be nice if I didn't have to buy an MRL tape just yet. If this is not the case then that's fine, I want to make sure I'm doing this right otherwise I think it's best to leave things as they are. Perhaps it would be nice to know of a simple way to check if it's relatively calibrated for 0db tape if this is the case and I'll leave it at that.

I do however have a true RMS multi-meter I could use to adjust the record levels and bias. I've found I like using non-backcoated Maxell 35-90, which is apparently the original tape the design engineers were using with the 388. In using the original tape, I'm hoping adjusting the bias pcb for 150mV as stated in the manual will work best. I don't see anything in the manual regarding using this particular tape stock so not sure why I've seen it mentioned a few times that they used 35-90, does anyone have any more info on this?

I'm having hard time understanding why the biasing involves measuring ac voltage at 150 mV and how this could relate to 0db bias tape.

the 35-90 bias specs are
Reference output level 200 nWb/m
0 dB bias Specified Bias of UD Tape
Playback EQ 3180 µs + 50 µs

So for example, say I wanted to adjust for -3b or +3db, would there be a formula for adjusting the ac voltage to compensate?

I think I'm mostly trying to understand the theory of how this is accurately biasing the machine before I start messing with it more than anything. I understand the procedure for a three head machine but this seems very different. Is the 1.5.6 Overall Frequency Response procedure where most of the calibration is actually happening depending on the tape being used, and the 150mV ac voltage is just a ballpark voltage for tape with a 0db bias?

I've searched these forums quite extensively but only see references from forum members wanting to bias for +3 and using the methods one would go with on a three head machine. If I've missed something or if there's some reading I could be doing elsewhere that would help simplify this for me please let me know.

Any help or advice from those who have navigated these waters before is greatly appreciated!

Thanks
 
Last edited:
"True RMS" digital voltmeters are almost all only 'true' at power frequencies.
Your very first step is to check the meter at 100 Hz, 1000Hz and 10kHz on tones out of a PC generated by (say) Audacity.

My Fluke 83 IS flat to 20kHz but no other meter I have tried is. The 83 was an expensive meter in its day. But, you would be far better off to find an analogue AC mV meter on the evil bay. Failing that you can build a very handy meter around a VU movement using a couple of op amps. Problem?

If such a relatively simple electronics project is beyond you then I would aver so is biasing and calibrating a tape machine.

Are you perchance in UK?

Dave.
 
" In using the original tape, I'm hoping adjusting the bias pcb for 150mV as stated in the manual will work best. I don't see anything in the manual regarding using this particular tape stock so not sure why"

If you want to mess around working on a tape that's out of production, I don't know if that is going anywhere. If you don't use a lot of tape, do whatever you want : )

One can, pretty much, hear when a formulation is close to matching factory (or, later re-calib) BIAS & EQ. To me, it sounds as if you going after this just messing around with a handful of tapes ?

Isn't there a 35-90 "replica" being made, today ? I've been meaning to sample that, but it is not needed as my little stock of 203 & 207 is good for 20-plus years as much as I use my 2-track & 4-track : )
 
...from looking at the manual, I've noticed I would only need a MRL test tape if I was attempting to calibrate the reproduction levels and bias.
Click to expand...

No. You use a calibration tape to set reproduce levels, reproduce frequency response and azimuth. You do NOT use a cal tape to set bias. You use whatever tape you are using to record to set bias.

The only checks ive done so far is to record at 0db and note that everything recorded at 0db appears to play back at 0db.
Click to expand...

Which doesn't tell you if your machine is calibrated to the outside world. The only way to actually calibrate your machine is to use a known reference source (cal tape, flux loop, etc.) You first calibrate the meters to known source tone level, then set the reproduce level using a reference source (cal tape or flux loop), and then from there you can set the record levels.

It would be nice if I didn't have to buy an MRL tape just yet.
Click to expand...

Then you also need to accept your machine may not be calibrated and you won't be able to calibrate it.

Perhaps it would be nice to know of a simple way to check if it's relatively calibrated for 0db tape if this is the case and I'll leave it at that.
Click to expand...

There is no cheat to find out if it is "relatively calibrated" for anything without a known reference such as a cal tape or flux loop. All you can do is calibrate your meters. You have no way of knowing if individual track record levels are high or low relative to your desired standard, and whether the corresponding reproduce levels are offset to compensate. The 388 is a 2-head machine so you can't monitor record levels on the fly, you can only monitor input levels and then reproduce levels, but buss 3 could show 0VU, be recording at -3VU and repro level set to +3VU to compensate. It'll look like 0 going in and 0 coming out but what is actually printing to tape could be all over the place. You need a cal tape if you want to move forward.

I do however have a true RMS multi-meter I could use to adjust the record levels and bias.
Click to expand...

As was stated above not all True RMS meters are equal. Most are only reasonably accurate around 400Hz, and grossly inaccurate not far from that. I have a Fluke 85 which is reasonably accurate across audio bandwidth. The 150mV is for Ampex/Quantegy 457...that's it. That's what the machine was designed for. I found, using the LF modulation method, and weighing that subjectively against the HF noise level, the bias setting for LPR35 was about 110mV IIRC. It will be something else yet again for a different tape type, particularly older "+0" tape. You're going to have to do the experimentation a leg-work using more traditional methods of setting the bias in order to determine what that level is. The method Teac prescribed in the 388 manual for setting bias is really a shortcut method and ONLY works relatively well if you are using THE tape for which the machine was setup originally according to the factory, and that's 457.

I've found I like using non-backcoated Maxell 35-90, which is apparently the original tape the design engineers were using with the 388.
Click to expand...

I believe that is incorrect...Ampex 457 is the tape from which the specs in the manual were derived.

In using the original tape, I'm hoping adjusting the bias pcb for 150mV as stated in the manual will work best. I don't see anything in the manual regarding using this particular tape stock so not sure why I've seen it mentioned a few times...
Click to expand...

Likely because in some regards it's easier to find being a consumer-grade tape, and no you don't see it referenced in the manual because that's not the tape designated for the machine. Who knows...150mV might work, but you don't know until you test it and check it out, and you CANNOT make assumptions in this if you want to do it right...457 and LPR35 are both 1mil +6 tapes, but require *significantly* different bias levels...I wouldn't assume 35-90 to be the same as 457 by any stroke until I verified that by setting the bias via a traditional method for the 35-90 and then measuring what the bias amp output is to achieve that proper bias setting. And if you're asking how I know about 457 being the factory specified tape, and why I went hunting for the proper bias setting for LPR35 and not just go with 150mV? Jimmy in analog support at Teac many years ago told me in a phone conversation LPR35 needed a much different bias setting, that the machine and the 150mV bias setting was for 457...I heard it from Teac staff directly.

I'm having hard time understanding why the biasing involves measuring ac voltage at 150 mV and how this could relate to 0db bias tape.
Click to expand...

Again, the 150mV measurement is a shortcut to make it easier for the end-user to bias the machine assuming you are using 457. It's an abnormal and *ineffective* means of setting bias...it does not actually set the bias for the tape you are using...unless that tape is 457. You have to use any one of a number of more traditional bias-setting procedures to bias a different tape. And the output standard of a given tape doesn't determine what bias level it needs...yes there are correlations. The higher the output tape, the thicker the oxide, and the more bias current needed to excite the oxide. But it's not +6 tape needs X bias level...refer to the differences between 457 and LPR35 for instance. And be careful...you referred a tape type as "0db bias tape". The output standard of the tape is referenced to nanowebers, not db. And the output standard of a tape does. It define a specific bias level needed. So there is no such thing as 0db bias tape. There is +0 output tape which, depending on the brand, specific oxide formulation, and specific run of tape, will require X bias level to be properly biased. The +0, 3, 6, 9 output standard designator on a particular tape is a reference standard represent at what level above the old +0 Ampex output standard the tape exhibits 3% distortion, and, again, we are talking fluxivity level...nanowebers/meter (nWb/m).

the 35-90 bias specs are
Reference output level 200 nWb/m
0 dB bias Specified Bias of UD Tape
Playback EQ 3180 µs + 50 µs
Click to expand...

This looks like a +3 tape...and the rest of the numbers just tell you what standard references were used when rating the tape.

So for example, say I wanted to adjust for -3b or +3db, would there be a formula for adjusting the ac voltage to compensate?
Click to expand...

No. And here again you are mixing db sound pressure level references when bias and output level has to do with fluxivity. Your sound pressure level adjustments are handled by your input and output level trimmers, but you can have 0VU going in and 0VU coming out but be hitting the tape with something more or less than that depending on how you want to use the tape...saturation levels, noise floor, etc.

I think I'm mostly trying to understand the theory of how this is accurately biasing the machine before I start messing with it more than anything. I understand the procedure for a three head machine but this seems very different. Is the 1.5.6 Overall Frequency Response procedure where most of the calibration is actually happening depending on the tape being used, and the 150mV ac voltage is just a ballpark voltage for tape with a 0db bias?
Click to expand...

Hopefully I'm not sounding like a broken record, but it is NOT a method for accurately biasing the tape, unless it is 457, and even at that it is just a close ballpark. And there is no "0db" tape...457 is a +6 tape.
 
Wow Sweetbeats! Good stuff! You have reminded me of a lot of the stuff I USED to know 40 years ago.

It had not penetrated this old noggin that the '150mV' OP was talking about was bias! Belay DMMs and the op amp meter I mentioned. Neither will be accurate at the 50, possibly 120kHz bias frequency and even if OP bought a milli V meter second hand I cannot see how he could check the calibration?

There is a 'quick and dirty' way to set bias, used for 2 head cassettes. Record 333Hz and 10kHz at neg 10 (-20VU for cassette) and tweak bias in steps to get the levels the same. Does not of course tell you anything about THD!

Dave.
 
ahh thank you sweetbeats. Everything is so much clearer to me now. I see now that +3b simply means that the tape (basically) distorts after +3b and how this have nothing to do with the fluxivity.

I do have some reels of LPR35 but never liked them as they sounded sort of brittle in the top end when saturating the tape.

Good to know that 150mV is for 457. Would be interesting to hear if anyone else has some recommend bias voltages for other tape stocks. Particularly what might have worked for them when using scotch 207 and 35-90. I'm guessing most 388 users just try a bunch of different tapes and see what they like best.

Well I suppose I need to get myself a test tape. The MRL 21T204 type seems to be the one I need judging from this chart. MRL Alignment Tapes from JRF

What is the simplest way of determining whether my multi-meter is up to the task of measuring the ac voltage? its a uni-t UT61E which from what I understand is about as good as it gets for a hobbyist meter.
 
ahh thank you sweetbeats. Everything is so much clearer to me now. I see now that +3b simply means that the tape (basically) distorts after +3b and how this have nothing to do with the fluxivity.
Click to expand...

You're getting closer, but no...it has everything to do with fluxivity. In the case of a +3 tape (not +3db or +3b as you last wrote...the label is just "+3"...), the test tone recorded to the tape at +3 (db) over the old Ampex standard, which equates to a fluxivity of 250nWb/m, results in a recorded tone that exhibits 3% harmonic distortion. That flux level that elicits 3% distortion from the recorded tone defines the Standard Operating Level (SOL) of the tape, and categorizes the tape as "+3", "+6", etc. So, yes, the sound energy level (dB, dBu, dBV, etc.) relates to the "+3", but the nuts and bolts of the identifier really relate specifically to flux level and distortion.

A +3 tape saturates at lower levels than, say, a +6 tape. This means you can get tape saturation with lower sound levels and stay within the headroom of your line amplifiers, and still achieve saturation. This is helpful if that's the sound you want but are using a narrow format machine like the 388...it is not usually preferred to switch off the dbx...but if you're using the dbx you have to be conservative with levels. So if you can't "bury the meters" to get some tape saturation without causing dbx tracking errors, your only other option is to use tape that saturates at lower levels like a +3 tape vs the factory specified +6 tape. The disadvantage with the lower SOL tape as compared to a higher SOL tape is the range between the inherent noise floor of the tape and the SOL is smaller on the lower SOL tape. This is the reason for the industry push to higher and higher operating level tape...not because it was hotter and had mo' bitchin' tape shiz, but because you could print hotter signals that weren't distorted and were also a greater distance sonically from the tape noise. But with the dbx on the 388 you mitigate the noise issue well if you want to use lower SOL tape for increased saturation...and no worries about line amp distortion.

I do have some reels of LPR35 but never liked them as they sounded sort of brittle in the top end when saturating the tape.
Click to expand...

And this begs the question: did you setup the bias for that tape? If you set it at 150mV, that is too much bias for LPR35...you will have increased distortion (brittle) and decreased HF response. Yes it would sound a little like a$$. I loved how LPR35 sounded on my 388 when I had one.

Good to know that 150mV is for 457. Would be interesting to hear if anyone else has some recommend bias voltages for other tape stocks. Particularly what might have worked for them when using scotch 207 and 35-90. I'm guessing most 388 users just try a bunch of different tapes and see what they like best.
Click to expand...

I've not read of anybody else doing what I did...like determining bias level using the LF modulation method or whatever. It was a PITA. I think most users just read the manual and assume 150mV works for everything without thinking through the logic of it and choosing to set the bias in a traditional way. It's a pain on a narrow-format 7.5ips 2-head machine.

Well I suppose I need to get myself a test tape. The MRL 21T204 type seems to be the one I need judging from this chart. MRL Alignment Tapes from JRF
Click to expand...

Yes.

What is the simplest way of determining whether my multi-meter is up to the task of measuring the ac voltage? its a uni-t UT61E which from what I understand is about as good as it gets for a hobbyist meter.
Click to expand...

What does your meter's specifications in its manual tell you about the accuracy of AC voltage readings? It should have specs regarding +/- % error rate at given frequency ranges. If you can't find anything like that in there assume it is not a meter suitable for audio bandwidth measurements.
 
"What is the simplest way of determining whether my multi-meter is up to the task of measuring the ac voltage? its a uni-t UT61E which from what I understand is about as good as it gets for a hobbyist meter." The specifications for that meter say it can measure frequency to 250megaHz! That does not however mean it will measure AC VOLTAGE with any accuracy at anything like that frequency.

Also the AC voltage spec starts from 220mV (-13dBV) and that is perilously close to -10dBu and so you would be reading playback output at the accuracy limit of the meter. You REALLY need a dedicated AUDIO test meter.

Checking the accuracy up to 20kHz is fairly easy. In a DAW (Audacity) generate test tones. Even a bog S internal sound card will deliver these with a flat response but make sure there is no EQ or fancy 'graphic' engaged.

Dave.
 
He can only reverse engineer a tape if his machine is first calibrated using a proper tape or flux loop. You can then track a tone set to a tape, but it would be a tape specific to his machine, and not as accurate as a proper test tape. I don't want the OP being misguided he can just make his own tape and it'll be the same as a proper test tape.
 
The old Ampex reference was 185nw,+3 is 250nw.The 388 was set at 250nw.Unless you are playing back tapes made on your machine on another 388,you should be ok on the playback alignment side.Your description of the LPR 35 sounding"brittle"could be because your 388 was biased for a different tape formulation.Usually screaming high end is because the tape is under biased- maybe the previous owner screwed it up.You can download test tones from a number of websites, a DAW or even a keyboard if you calculate the freq. needed.Record 100HZ,1KHZ, and 10KHZ [MENTION=16808]0vu[/MENTION] with the DBX off.Do they come back close to zero,(-1db), @ 10KHZ relative to 1KHZ? Yes,don't mess with it.No, means you will have to fine adjust the bias.That 150mv gets it close,but fine tuning each channel to playback the 1KHZ/10KHZ at the same level is the easiest way to go. Remember, every time you change the bias on each channel,you will have to do the record level calibration on that channel. Don't get in there and turn things,(fingeritis), for the sake of turning them-that's how I made a lot of money in the past on 388s :-0 .Biasing is best left to someone who knows what they are doing,and if you are going to do that,you might as well pay the tech-verify he has the correct MRL or Teac playback alignment tape,along with a Tentelometer to do the whole alignment including tension and speed calibration,along with input and meter cal. You may find the tech will recommend re-lapping the R/P head.A groove worn in the head the width of the tape -.25"- means it should be done if you plan on keeping and using the 388.Supply the tech with a reel of the tape you will be using- everytime you change tape formulations you will have to rebias,so stick with one formulation.Running the 388 at any elevated level,(+6 - +9). will create huge amounts of crosstalk-"splatter"-because of the narrow track width and having all eight tracks on one head-inductive coupling between coils.Use as -is with a few quick tweaks,or get it done right.
 
Making a copy of the test tape will not work.The playback-MRL- alignment tapes are made "full track width". This means there is only one track on any MRL up to 2",and the phase-head azimuth- is the same across all tracks.You can't duplicate that with a multi track head.
 
I gots plenty of full track - none have a needle meter , but all you need is something calibrated
 
garww said:
I gots plenty of full track - none have a needle meter , but all you need is something calibrated
Click to expand...

Hey, stepping out of the shadows for a sec. I don't mean to be rude but it's foolish to disregard the technical advice being given here. The people commenting on your thread have decades more experience than you do and they're doing their best to help you with your situation.

And, as someone who's aligned numerous machines myself, I can say they're right, and you need to listen to them.

Now.... smile! Buy a MRL, follow the correct procedure for aligning a tape machine, and be thankful afterward that you didn't mess up your 388. :)
 
The guy to listen to here is sweetbeats.
There is nothing to doubt/question or sidestep with his explanations of how to do it and/or what is needed.
 
You must log in or register to reply here.
Back
Top