DIY Alignment and Calibration

Alignment tape

Ok, so after months of disgust, I'm getting back into this project of tape deck alignment for my tascam 38.

Last year I purchased a rmgi sm911 tape which I thought was compatible with the old ampex 456 tapes.

I was searching online looking to buy a new mrl reference tape. The tascam manual says to use a 15ips, iec1, 250 n/wb tape. I'm good to go so I thought. So I stumbled upon http://usrecordingmedia-store.stores.yahoo.net/ancaltap1.html and that page says to use 355 n/wb alignment tapes for the sm911.

I thought I had a handle on this. Can someone offer a clarification?
 
fstrat, guess who asked that very same question? I'll give you a hint: go here and read that thread.

[I'll give you another hint...it was ME!]

Read that thread. I hope it helps, but the biggest thing to get is that the output rating of the tape you use (+3, +6, +whatever), is not related to the operating level of your calibration tape or what level or standard you set your machine for.

The +3, +6 +whatever of the tape is a (get this) measure of how hot a level the tape can handle before exhibiting 3% distortion. The higher the rating, the more powerful a signal a tape can handle before distorting.

The strength of the signal can be measured in lots of ways but when we are calibrating we are working with electrical measurements (dBv's, dBu's, etc.) and magnetic measurements (i.e. "flux" measured in nanowebers per meter, nWb/m).

Electrical measurements measure the strength of the signal in the cables and in the electronics, magnetic measurements measure the signal strength on the tape.

Read that thread.

Let me see if it ties it together for you.

Let us know. It took me several times walking through the terms and the process to really understand but you can use any flux level cal tape to set your deck up. You just have to compensate what your meters read from what comes off the cal tape.

You're good. RMGI SM911 is 456 compatible. Just right for your 38, and the 250nWb/m tape is the one you want. Tascam spec'ed the 250nWb/m cal tape (which is equivalent to "+3") because the electronics were designed around using 456 tape ("+6" tape). If you use the 250 nWb/m tape and zero your meters using that "+3" flux level, then your meters are going to be hitting zero with "+3" signals...you've still got 3dB's of headroom with your "+6" tape before it starts distorting. This is a simplified overview because of course setting the deck up with tone in a calibration process is different than how the meters and the tape respond to real-world program material, but we are simply setting the deck up to a standard so all the channels/tracks behave the same and then you let your ears judge! ;)

Once you get a grip on this (and we'll help), then a whole world of options opens up and you start to understand why people set their decks up to different standards and select tape based on what kind of sound they want on playback...you'll join the legions of analog die-hards that understand the foolishness in getting the hottest tape and taking everything into the red to get that "phat tape compression" when (depending on the program material and the machine and signal path) they'd have much sweeter results leaving some headroom in their electronics and using "+3" tape like 406!! Yes, its really true. :)
 
Sweetbeats, thanks much for the great explanation and the other thread link. I read it all. Believe it or not, I think i got it!!

I read some of the mrl site today as well and understand about the compensation needed if I were to get a different flux level tape but I will just stick with the 250 nWb/m in the 38.

I use a lot of daw so I might want the 38 to print a little hotter but I will start at the spec level first.

Looks like some fun coming up as soon as I get the tape.

Thanks again!
Mike
 
Mike, yup! Starting with the factory recommendation is a great place to start and then you can explore from there and a 250nWb/m cal tape will serve you well.

Good luck, have fun, and post back to this thread or start a new thread as you journey along.
 
OK, so I got all the tools, alignment tape, etc. I've started this process and the first thing to do is a head alignment check.

I put Winscope on it (tracks 2 and 7) and it looks to be off about 30 degrees. I was hoping it would be aligned but considering it was banged up in shipment to me, I guess I'll have to attempt an alignment.

My question is: The manual does not say to scope both heads but rather talks about the gaps between the heads.

Do I do the scope for the sync head, adjust it, then the repro head is another procedure?

There's two hex adjusting screws per head. How do I know which side to start with?

Not quite understanding what to do here. A little nervous about ruining this.


Mike
 
I put Winscope on it (tracks 2 and 7) and it looks to be off about 30 degrees.

Just so everybody is clear, we are talking about adjusting the azimuth which is making sure that the vertical axis of the headstack is perpendicular to the axis of the tape travel.

Tascam spec'ed azimuth to less than 90 degrees which would look like a circle on the scope. Now, do I suggest you be happy with 89 degrees? Nope. And you aren't happy with 30 degrees out and I don't blame you. Just making the point that, according to the service manual, 30 degrees is within spec.

My question is: The manual does not say to scope both heads but rather talks about the gaps between the heads.

Do I do the scope for the sync head, adjust it, then the repro head is another procedure?

Typically manuals I've read will have you adjust each headstack separately while reproducing the azimuth adjust tone on the test tape (usually something like 10kHz for initial adjust and then something like 16kHz for fine adjust)...IOW you'd do a complete azimuth adjust procedure for the sync head and then when you're done with that do the same procedure on the repro head. I think that's fine, But I'm going to be doing it as its implied in my Ampex MM-1000 manual which is to do the procedure on the sync head, and then record tone onto tape and reproduce using the repro head and than adjust off of the tone recorded to tape by the sync head. In theory this should be the same as doing the procedure separately using the test tape, but the key goal is to have the repro head as close as possible in azimuth to the sync head, so I agree that the sync head should be the reference for the repro head, not the test tape. This will take any anomalies of the transport into account. Again, in theory it shouldn't make a lick of difference, but why not take a middleman out of the procedure for the repro head adjust? Anyway, my $1,000,000 disclaimer is to do it like it says in your manual not like "that SOB sweetbeats told me to do it and now my deck's all wonky..." :mad: :eek::D

There's two hex adjusting screws per head. How do I know which side to start with?

Not quite understanding what to do here. A little nervous about ruining this.

Don't sweat it. Your heads are screwed on tight to the headblock with a larger screw from underneath...the azimuth adjust hex-set-screws just kinda tweak it one way or the other. In theory the head should be perfect azimuth without the set screws being used but Tascam knew that would not be a good idea to assume that. Just start by making sure that both screws are bottomed (i.e. not loose), not *tight* okay? But so that each screw is contacting the headblock. Check this while watch the scope and you'll see and get a feel for what kind of tweaking does what, and if tweaking one screw makes the oval get more roundy-circle-like then back off on that one...you'll realize a point when you are backing off that the oval doesn't get closer to being a line (i.e. 0 degrees) because the screw you're tweaking is now loose and that's when you need to tighten the other screw, but you want the that screw to have to push against the first screw so that once you find the right position it is held fast in the position. Does that make sense? If both screws are pretty snug but the azimuth is not yet where you want it, then you loosen one a little bit and tighten the other. I hope this helps and isn't more confusing. I think if you make minor little adjustments with the tone reproducing while you're watching the scope it'll make sense, and the way the Tascam heads are on the 1/2" 8-tracks your worst-case-scenario is to start fresh and back both set screws off...that will likely leave the head sitting at less than 90 degrees out of adjustment so you're not likely to mess things up to where you can't find your way back.
 
OK, thanks for the info. So far so good. I understand everything, scary.

So I've started, no turning back now. There goes the summer.

Before azimuth adjustment, I decided I wanted to check (since I purchased this deck used) if there was any permanent magnetism on the heads. I demagnetized with han-d-mag.

I followed instructions here at MRL: http://home.comcast.net/~mrltapes/accurate.html

For fun, I put a 1kHz tone on track 5 and recorded at 0VU for about 30 seconds. Playback on the sync head looked good, same level, but the needle fluctuated just a little.

I then recorded a 20kHz tone on track 5. While track enabled, no other VU meters were moving, perfect. I put it into record and the adjacent tracks (4 and 6, not record enabled) VU meters were picking up the signal about 7db less that the VU on track 5. This did not occur with 1kHz tone. I continued to record the tone to track 5 anyway.

1) Is it normal for such crosstalk between tracks?

2) More importantly, there was a 2db drop in playback, but each subsequent playback, about 10 playbacks total, yielded the same -2db result, so I think I'm good to go as far as not having any permanent magnetism, otherwise I should have seen db drops on each playback, right?

3) However, the playback on the VU meter fluctuated, meaning while it mostly played back at -2, the needle would move slightly occasionally (unlike when I recorded it at a steady 0 VU). Is this normal or will this get corrected as I proceed through the electronic adjustment steps?


This place has been great and very helpful and I appreciate the input and help!

Learning a bundle.


Thanks,
Mike
 
1) yes. It is not an actual crosstalk between tracks but signal induction in the meter circuits IIRC. Dealt with this exact same thing and reacted as you did with my 58. Its okay. You'll notice that if you arm the tracks (4, 5 and 6), monitor the input (your 20kHz tone) but only route the signal to track 5 as you have been doing, if the adjacent tracks are record enabled the meters of tracks 4 and 6 don't react, but if you disarm tracks 4 and 6 then the meters act up. This is notmal. You have to realize that the penetrating energy of a 20kHz tone at 0dB is pretty stinkin' strong.

2) I don't know about this...where did you get that info because that would be good for me (IOW I'd like to know) if that is a valid test for permanently magnetized heads...that's be cool if it is as I've never known a way to test it.

3) How much movement? I was worried about this too on my 58 but some movement is to be expected especially on edge tracks and at higher frequencies. It can be related to minute changes in tension, maybe older tape (is it new tape or have you bulk erased it?), etc. The rule of thumb I was given was that even fluctuations of as much as +/- 0.5dB can be expected. You are talking about the meter needle wavering during playback right?

You're doing great. And really, don't sweat the azimuth adjustment. Zenith is the really hairy one (the face of the head being parallel to the surface of the tape), but lucky you the zenith is fixed on the 38. You'll be just fine. You strike me a studious and consciencious operator. You'll be fine.
 
fstrat,

I found the dialog on the meter crosstalk issue related to me 58.

Here's what Beck said in this post:

METER CROSSTALK ISSUE
Some interaction is to be expected while in sync mode at high frequencies and it only affects the adjacent meters, not what goes to tape. The higher the frequency, the greater the deflection at a given level. The phenomenon varies depending on the machine.

The first thing to keep in mind is this won’t cause any problems during real world recording… at least it shouldn’t if there’s nothing wrong with the machine. This is because when recording music the only thing that lives near 20kHz are upper harmonics and they are nowhere near what you’re hitting your meters with in your test.

A couple other tests you should try to confirm this:

1. Perform the same test with track 6, but also set record enable on tracks 5 & 7. Make sure you only have signal going to track-6 just like before. At 20kHz the deflection on the adjacent meters should be much less (maybe none at all).

2. Also do the same test with music from a CD or other source instead of a test tone. With music peaks around 0VU the adjacent meters should not be moving.

3. If you’re still seeing more interaction between the meters than you should, try connecting the outputs from your mixer to the corresponding inputs on the deck (except the input from the oscillator, of course), then perform the tests again. When you have the inputs (or outputs) disconnected (floating) it may allow spurious signals to affect neighboring components such as adjacent meter circuits. Even with all the faders down on your mixer the circuit for each channel is still complete and more resistant to interference.

Also keep in mind that recording a high frequency test tone while playing back another on an adjacent track has no real practical utility. You want to record each track and then play them back together (or one at a time) to get a picture of the machines frequency response. So while the meter interaction may be annoying it’s only a problem if it occurs while overdubbing during multitrack recording of music.

Should put your mind at ease.
 
sweetbeats,

1) For the meter crosstalk, that's great news. I will test with 4 and 6 record enabled. Yes, I didn't even think 20kHz at 0 VU is strong which explains why a 1kHz tone did not affect tracks 4 and 6.

2) The 20kHz test info I got from a section under magnetism at the http://home.comcast.net/~mrltapes/accurate.html link which I researched from Beck's post at https://homerecording.com/bbs/showthread.php?t=241577 and https://homerecording.com/bbs/showthread.php?t=226932 post #4.

3) Yes, the meter wavers during playback, I would notice more than .5 db on the same places on the tape. A couple of places where it was 1 or 2 db dips. The tape is a 6 month old RMGI 911, recorded only on as a test a few months ago. Perhaps electrical adjustments will correct this or maybe even the tape is defective.

I will do a couple more tests on the magnetism procedure then move on to the azimuth adjust next.

Good stuff, thanks!

Mike
 
1) Cool!
2) Ah! Thanks for putting those links together. Excellent and relile sources each one of them. Thanks!
3) Okay. Interesting. I'd say if you have a way to bulk erase the tape and then track tone to it again in those spots and see if you still get dropouts. I wouldn't worry about it really but if that happens throughout the tape I'd be contacting the dealer and seeing what they say. Who did you buy the tape from?
 
Got a quick side question...

Without looking through my Otari manual (I'm not at my studio at the moment)...last summer I calibrated my 2-track deck using a NAB EQ MRL tape.
That is what John French from JRF Magnetics recommended for that deck.

Of course, the deck can do either NAB or IEC...and it seems a lot of folks prefer the IEC (my 16-track uses IEC).

So, I was reading this thread and started wondering how many of the settings I would need to redo if I decided to switch from NAB to IEC on that deck.
I don't need to touch any of the mechanical adjustments...and I'm thinking I would only need to redo a couple of the electronic adjustments...but like I said, without the manual in front of me, I don't know for sure.
 
3) Okay. Interesting. I'd say if you have a way to bulk erase the tape and then track tone to it again in those spots and see if you still get dropouts. I wouldn't worry about it really but if that happens throughout the tape I'd be contacting the dealer and seeing what they say. Who did you buy the tape from?

I don't have a bulk eraser, but if it continues to happen I will check with where I bought the tape, which was usrecordingmedia.

Now that I'm really getting into this, I'm wondering about the transport and such which I understand should be in order before other adjustments. I have two items which I don't think are major issues, but think I better check on this before going any further:

1) All my reels move up and down on the reel table. None of the reels I use scrape the tape but since I have two new reels (the rmgi tape and the mrl tape) plus a new empty reel, they all exhibit this behavior. Height seems ok but I am wondering about the levelness or adjustment of the reel tables. Is this typical for the 38 or are the flanges just never that flat (hard to tell by looking or even with a straight edge)?

2) I noticed that when I press play when the tape is stopped and positioned over recorded material, I get a very, very minor short screech of audio as if the transport is pulling the tape across the heads before it fully kicks in.

Should any of these warrant looking at the transport first before proceeding with azimuth and electronic adjustments?
 
1) If the tape isn't coming in contact with the flanges you're fine. Standard reels (3-screw) with standard thickness flanges commonly display undulations as they rotate. Its one reason I like the precision 6-screw reels with the thicker flanges as they tend to rotate more true. The other and more important reason I like the precision reels is that the hubs fit more snugly on the reel adapter and thus display less eccentricity when rotating. Just remember, your 38 has a complex servo tesnioning system designed, in part, to deal with all these issues and, like I said above, if the tape is not hitting the flanges I'd leave well enough alone.

2) This should be fine.
 
Progress:

Azimuth adjustments are complete, no big deal at all really, although I was surprised that the screw adjustments weren't as "fine" as I expected, meaning it wasn't as easy to turn the screws. The only issue I ran into was that WinScope couldn't track anything useful over a 4kHz tone, even with a lot of gain into the PC. I was able to adjust all frequencies and lower. Is this a problem if I don't adjust azimuth at much higher frequencies? All others look really good.

Moving on to the electronic adjustments, I went through these procedures twice just to make sure. Interesting to note that on a 16kHz Repro and Sync EQ adjust, to reach 0VU, tracks 1, and sometimes 2, would take their time reaching 0VU, but all looks pretty good.

Now on to the next set of adjustments, bias. However, there's a problem here in the 38's manual. The first picture in the manual of the pot locations (page 36 in my manual) shows the location of C134 Bias as #8 as the first left most trim pot, with L106, bias tuning, as the second. However, pictures on pages 37, 41, 43, and 83, show C134 as being the SECOND leftmost pot.

I'm guessing its the second one, not the first.

Also, I'm a little confused that it says that if there is not enough clockwise motion available to reach peak before a 4 or 5db drop, to adjust the output level of the oscillator. OK, I understand that, but then it says "if you have moved the input level pot on the front panel of the 38 to keep your reading on scale, the next adjustment will correct your input reference". Huh? What level pot on the front of the 38? That and the next paragraph is a bit confusing.


Almost there I think.
 
Azimuth adjustments are complete, no big deal at all really, although I was surprised that the screw adjustments weren't as "fine" as I expected, meaning it wasn't as easy to turn the screws. The only issue I ran into was that WinScope couldn't track anything useful over a 4kHz tone, even with a lot of gain into the PC. I was able to adjust all frequencies and lower. Is this a problem if I don't adjust azimuth at much higher frequencies? All others look really good.

Hm...4k is pretty low for azimuth...10k, 12k and even 16k are really preferred for azimuth. You're probably fine but higher would be better. What are you using as an audio interface? You might want to try True RTA. It has a built in scope function, is free and I really like it. But, bottom line, it wasn't as scary as you feared was it?

Moving on to the electronic adjustments, I went through these procedures twice just to make sure. Interesting to note that on a 16kHz Repro and Sync EQ adjust, to reach 0VU, tracks 1, and sometimes 2, would take their time reaching 0VU, but all looks pretty good.

Can you clarify? What do you mean by "take their time"?

Now on to the next set of adjustments, bias. However, there's a problem here in the 38's manual. The first picture in the manual of the pot locations (page 36 in my manual) shows the location of C134 Bias as #8 as the first left most trim pot, with L106, bias tuning, as the second. However, pictures on pages 37, 41, 43, and 83, show C134 as being the SECOND leftmost pot.

Trust what you see on pages 37, 41, 43 and 83 and the trump is to just look at the card. The coil L106 as well as C134 will be labeled as such on the PCB itself. You can also look at page 102 which has the PCB layout and you can see that it goes (from left to right) TP-3, TP-2, L106, C134, etc. Gotta love those Tascam typos. These older full manuals are GREAT manuals but they are not without their errors too. Good catch on this one.

Also, I'm a little confused that it says that if there is not enough clockwise motion available to reach peak before a 4 or 5db drop, to adjust the output level of the oscillator. OK, I understand that, but then it says "if you have moved the input level pot on the front panel of the 38 to keep your reading on scale, the next adjustment will correct your input reference". Huh? What level pot on the front of the 38? That and the next paragraph is a bit confusing.

What page is that on? Don't rack your brain about it unless you have a problem having enough trim range to get to the proper drop point. What tape are you using? AFAICT All they are saying is that if you change the level of the source then you'll want to adjust the meter trim or input level trimmer so that "0" still shows as the level of your tone coming off the oscillator. Kind of silly IMO since you use the meters to show relative changes (i.e. you're not trying to get the tone to "0" or something...you note the level of the peak as you turn the trimmer clockwise and then quit turning when you reach 4~5dB's below that peak, whatever it is...doesn't matter the VU value), and after I cal my input levels and meters I wouldn't want to mess with it again anyway.
 
Hm...4k is pretty low for azimuth...10k, 12k and even 16k are really preferred for azimuth. You're probably fine but higher would be better. What are you using as an audio interface? You might want to try True RTA. It has a built in scope function, is free and I really like it. But, bottom line, it wasn't as scary as you feared was it?

I was using my laptop sound card line input, and now that you mention it, I should probably use a real audio interface. The onboard sound card probably is not the best suited for this kind of work.


sweetbeats said:
Can you clarify? What do you mean by "take their time"?

For the 16kHz Repro and Sync EQ adjust, the meters on tracks 1 and 2 "take their time"; increasing to 0VU slowly rather than immediately like the other tracks once Play is hit. Its odd, so maybe something is wrong here?

sweetbeats said:
Trust what you see on pages 37, 41, 43 and 83 and the trump is to just look at the card. The coil L106 as well as C134 will be labeled as such on the PCB itself. You can also look at page 102 which has the PCB layout and you can see that it goes (from left to right) TP-3, TP-2, L106, C134, etc. Gotta love those Tascam typos. These older full manuals are GREAT manuals but they are not without their errors too. Good catch on this one.

Thanks, I was just concerned turning the wrong pots and mucking everything up.

sweetbeats said:
What page is that on? Don't rack your brain about it unless you have a problem having enough trim range to get to the proper drop point. What tape are you using? AFAICT All they are saying is that if you change the level of the source then you'll want to adjust the meter trim or input level trimmer so that "0" still shows as the level of your tone coming off the oscillator. Kind of silly IMO since you use the meters to show relative changes (i.e. you're not trying to get the tone to "0" or something...you note the level of the peak as you turn the trimmer clockwise and then quit turning when you reach 4~5dB's below that peak, whatever it is...doesn't matter the VU value), and after I cal my input levels and meters I wouldn't want to mess with it again anyway.

This is on page 41 of the manual, top right and continues on page 42. The re-adjustment of input level is what is confusing here. I have not done this step yet but will give it a whirl this weekend.


Thanks for the help. Much, much appreciated. I will summarize the procedure in this thread as soon as I can get through the entire procedure.
 
For the 16kHz Repro and Sync EQ adjust, the meters on tracks 1 and 2 "take their time"; increasing to 0VU slowly rather than immediately like the other tracks once Play is hit. Its odd, so maybe something is wrong here?

Can't recall how the channels are paired up on the cards...channels 1 and 5 on the top card or 1 and 3? Anyway, try swapping the card that has channel 1 on it with another card (that doesn't have channel 2) and see if the problem follows the card.
 
Can't recall how the channels are paired up on the cards...channels 1 and 5 on the top card or 1 and 3? Anyway, try swapping the card that has channel 1 on it with another card (that doesn't have channel 2) and see if the problem follows the card.

I finished up the procedures, started over everything; tweaking the azimuth again since I was finally able to get 8kHz, 10kHz, 16kHz to show up on the scope for azimuth settings. Still, the azimuth screws don't have a threading that allows for "fine" adjustment, but I'm almost dead on here anyway.

The Bias adjustment was tricky watching for peak and 4-5db drop.

The L103 adjustment for the last step Peak Adjust didn't seem to do much of anything, but all tracks seemed like they were in spec anyway.

The bad news is I think I have a bum track/channel 1. I can't seem to get high enough readings on high frequency tests 8kHz 10kHz 16kHz 20kHz. When recording a high freq tone and checking the VU on the repro head, the VU is very low, then increases slowly to normal, then drops back down to nothing, up slowly again, on and on. Repro cal with the test tape are dead on and look good. Input cal is perfect also. Unless it could be the tape, I am going to have to swap the channel cards I think.

The problem is getting the cards out. Channel 1 is the top card, 2 is next and so on. I removed the brace holding the cards, but can't seem to "easily" pull the channel card out. I don't want to break anything here.

Is there something else holding these cards in place besides the one brace?
 
Back
Top