Levels Recorded To Tape Hotter Than Input

M

mjiggidy

New member
Hi! I'm very new to analog recording so I'm sure this is a very beginner question. Thanks in advance for the hand-holding.

I have a TEAC A3440 (1/4" 4 Track machine) using SM911 tape (purchased new, Pyral branded). I'm feeding it four tracks of line-level signal from a Focusrite Scarlett 18i8 that I want to record to tape.

To calibrate levels, first I send out a 1kHz -20dB tone to each of the four tracks, and set the TEAC's on-board meters to 0 VU. This fine, but once I start recording to tape, if I switch the TEAC's monitoring over to "playback" so I'm monitoring off the tape, the same tone is read as something like +5 or 6 VU off the tape for each track. During record, I can toggle between input monitoring and tape monitoring and see the levels from the tone are still being received at 0 VU, but are coming off the tape at that +5 VU.

Of course, I can just start recording first and re-adjust until the tape monitoring levels to say 0VU and it's fine. But this difference in levels still puzzles me.

So here's where I'm dumb. I hear people talk about tape bias, and although I never totally understood what that was, I feel like maybe that has something to do with this? Am I over-biased? Am I way off?
 
Last edited:
The machine has almost certainly been set up for a differnent tape formulation. My guess is that it was set up for 406 or something.

Different tape formulations have different output levels for a given input. For example - and these are purely made-up numbers - Quantegy 406 might take 15v to record a signal and when you play it back, you get 1v coming into the head. If you were to record on ATR tape, which is higher output, 15v in might return 3v in playback so if the machine was calibrated to read 0VU for 1v, it'll read a far higher level on playback. Again, these are totally fake numbers... it's probably microvolts on playback.

Bias is also related to this, but it's frequency-dependent - in our example, a 100Hz tone might record at 15v and return 2v, while a 1Khz tone returns 1v, and 10Khz returns 0.5v. Adjusting the bias will change these numbers and the trick is to try and move it around so that they all return 1v, but it's a compromise since increasing the high-frequencies will suck at the low frequencies. It also causes distortion. If the bias is too low you get this weird sound like rocks being scraped around. If it's too high it will lose the high frequency response.

What you need to do is get a test tape. That machine is probably intended to run 456/911 on 250nWb/m but that's just a guess. First you adjust the playback levels until your 1Khz tone recorded on the test tape reads 0vU on all meters. You'll also need to ensure that 100Hz and 10KHz read 0 as well, which may require adjusting the head alignment and possibly the EQ trimmers if there are any, and then go back and make sure that the 1KHz playback is still okay.

If the deck supports sync playback you will need to do the same thing for the sync head as well. Now your machine is set up play back at the proper level. Then, and only then, can you set up the record levels.

You can get an approximation just by recording 1KHz on on your fresh tape (Not the test tape!), monitoring on the playback head, and adjusting the record level until it matches 0VU on playback. But to do it properly you'll also need to do the same for 100Hz and 0KHz and adjust the bias level until you get a reasonable compromise.
 
Thank you very much for the informative post!

So if I'm understanding correctly: the "important" reading where I want to be sure to see 100Hz/1kHz/10kHz tone at 0 VU is from the playback head, correct?

If I do want to set up my machine correctly so that the input levels match the output levels, is that something I can do with the test tape? Or do I need to have it serviced? Is this process what "head alignment" is?
 
Yes - 100/1000/10000 are the usual test frequencies. Any multi-frequency test tape should have at least those three, though they may also have others, e.g. some people test at 16KHz as well, and some like to test at 40Hz because low E on bass is around that point.

It is worth bearing in mind that the frequency responses are never flat on analogue machines. While the manufacturers ensured that they were flat at those three frequencies, they can go completely weird in between. There's some interesting graphs here: Response Curves of Analog Recorders

But yes, you really need a test tape to set the machine up properly, because that's an outside reference. If you just rely on a tape that you've recorded yourself you'll end up chasing your tail because you have no idea if the playback or record side is out of spec.
There are two kinds of alignment - head alignment, which is where you make sure that the heads are properly adjusted mechanically. For that you need an oscilloscope and preferably the test tape although you can do some degree of alignment with just the 'scope and a test tone. That will ensure that the record and playback heads are in phase with each other.
Electronic alignment is the other one, which is what I was describing earlier. That's where you play back the test tape and adjust the trimmers for each channel until they read 0VU. (And then the sync head, and then the recording levels and bias as well).
See if you can find the manual for the machine - the service section it will have information about setting up the machine in this way.
 
Thank you once again for your amazing posts! You're making about a million things click in my head that I wasn't understanding before!
 
The deck might not of been set-up for anything - just needing the equiv. formulation of the recommended tapes
 
That is a point. I figured calibrating the machine for SM911 would be the way to go, but it you can get 406-compatible tape it might be worth trying the machine with that and leaving it alone.
I think that Splicit Capture stuff is roughly compatible with 406, but I'm not sure... alternatively, if you can find old-stock Quantegy 406 or 407 on ebay, maybe try that. Don't use the Ampex ones, it's likely to be sticky.
 
I think he needs to confirm that it's actually a level boost on playback... and not just the meters needing to be calibrated for both input and output match.
 
I'm fine with unopened 203 and 206 for the two machines in my music room, but I dunno what was recommended for the 3440. 203 was recommended for the old 2340, and the other had a sticky label for 406 - a ampex version of the 25-2. I would think (hahahah) the 3440 would be good with +3, or, +6 ???
 
I'm using a TEAC A-3340, which should be very similar to your later machine.

The manual indicates a number of different tape types as being 'suitable', and I regularly use various types myself, some are on the manual list, others are not. I have some of the SM-911 on order, and another machine I have (a Ferrograph Logic 7) has recently been serviced and specifically 'tuned' for this tape.

I note that some tapes do play back slightly louder that they were recorded (as per VU meter settings), but I don't really worry about it. I listen out for any hint of distortion, and act accordingly. My understanding is that some tapes can cope with higher levels, and +3 for one tape could be perfectly fine, while another tape SHOULD be kept within 0 VU. Note, by the way, that SM-911 is a SP type, so the tape is slightly thicker and therefore has more magnetic coating and could well be slightly more retentive/sensitive that LP types anyway. That aside, I have a spool of Racal-Zonal 889, that is LP, but I certainly note that this tape shows a 'tape' level slightly more than the 'source' level, and I've sometimes had the peaks hitting the max on replay, but the sound quality is still excellent. No distortion. So the tape clearly CAN copy with the higher levels.

If you are planning to use the single tape type, then there will be benefit to doing what is being discussed, so that all your level indicators are consistent, but tape is not always so easy to get hold of these days, so you may sometimes use whatever you can get.

Geoff
 
406 and many are +3 tapes. On my 2-track one three position switch reduces level to the meter, so it stays at 0 level. So, I should be able to meter normal, +3, and +6 tapes

And there are hotter formulations
 

Attachments

  • eq_tape.jpg
    eq_tape.jpg
    29.4 KB · Views: 3
Like everything else on a recorder, meters need to be calibrated. TEAC machine meters are frequency weighted. The only frequency that they are accurate at is 1Khz.
 
You must log in or register to reply here.
Back
Top