How to make a Speed Calibration tape

[Beck]

Can’t make a speed tape without a known reference, in the same way you can’t make a calibration tape without a standard reference. The tone is a precise 3.15 kHz as recorded at MRL on machines with very tight tolerances.

The best deal in town is the TEAC YTT series that have the speed/wow & fluter tone on the regular calibration tape. For ½” tape it’s model YTT-1144-2 or the latter YTT-1144SP. The original YTT-1144 doesn’t have the speed tone and is vulnerable to sticky-shed.

Otherwise you have to buy a regular cal tape and the speed tape from MRL.

Trying to make your own without knowing where your deck is calibration wise would be like trying to construct a scale from scratch without a known reference weight.

 

[Dr ZEE]

...motors should run at 15 ips....

from where to where?

 

[Dr ZEE]

Can’t make ...

well, Tim, the theory here is pretty good. The practical side is a diffrent story.

If you "mark" the tape periodically and know the exact "distance" between marks (pulses , that is) and you have a device that can "register" the "frequency of marks" while the tape is running, then you can adjust the tape speed to what ever is desired.

If you drive on a straight highway and the highway has marks on it every known distance (period) (say, each 1 meter) , and your car has a "device" that "looks" down and can register marks and calculate the "frequency of the marks" (the amount of marks passed per period of time) , then you can figure out the speed you are moving and so you can adjust the speed to what ever desired
Ethan's "method" is not to figure out the exact distance between marks (pulses), but amount of marks in a known length of a path (which is the same thing )

 

[sweetbeats]

The best deal in town is the TEAC YTT series that have the speed/wow & fluter tone on the regular calibration tape

Tim...so...wait...that 1144-2 I bought from you is also a speed/W&F tape??? I should know that...I've run it on my 58 bunches of times but have pretty much used the first half of the tape with the bias and response tones...haven't really "listened" to interoductions to the tones later on the tape...plus its been awhile since my 58 has been out of commission for...wow...awhile.

If that's the case then that IS the best deal in town!

 

[jpmorris]

Couldn't you just use any old test tape and a frequency counter?

 

[sweetbeats]

Hm...good point...you mean any old test tape with a 1k tone on it right (which any old test tape would have or should have AFAIK...)?

Is there something different about the tone on the speed test tape?

 

[evm1024]

Sure

Couldn't you just use any old test tape and a frequency counter?

Of course - However MRL says that the alignment tapes are not set up for speed adjustment. But I'm sure that they are within a few percent.

One would need a calibrated counter in any case.

The interesting thing about making your own tape is that you have a relitivly well known accuracy. Might be useful when replacing the motor or when you suspect that the speed is off for some reason or other.

Regards, Ethan

 

[miroslav]

Maybe this will work....and it's cheaper than the test tape:

http://www.opticsplanet.net/busspeedrads.html

Of course...you would have to then do the mathematical conversion from MPH to IPS....



Sorry...I couldn't resist.

 

[sweetbeats]

I want to just weight in on a serious note for a sec...there have been a few posts that seem to be indirectly pointed to the OP that sort of state "why would you do that/go to all that trouble, etc."

Always used to be that you had to have the spedd cal tape to do it...now Ethan has come up with a way to do without the speed cal tape which, in my book, is really cool and a part of what this forum is for. Yes you still have to have a good pulse counter and frequency counter, but ultimately it is an alternative that might actually work for somebody and (at least for me) has been interesting and educational to boot.

Thanks for sharing the daydream.

 

[Beck]

Tim...so...wait...that 1144-2 I bought from you is also a speed/W&F tape??? I should know that...I've run it on my 58 bunches of times but have pretty much used the first half of the tape with the bias and response tones...haven't really "listened" to introductions to the tones later on the tape...plus its been awhile since my 58 has been out of commission for...wow...awhile.

If that's the case then that IS the best deal in town!

Yep! It's the last tone on the tape -- 3.15 kHz. It's just a standard. People are asking, why not any other tone? To be in the ballpark another tone may be fine, but like everything else standards were developed so machines in every studio would be running at the same speed. By the way, there’s no spoken introduction on that tone for some reason… it just starts. Don’t know if there’s some reason to keep other tones or spoken print away from it or what.

Another thing to keep in mind is that speed changes ever so slightly on an open-reel machine as the ratio of tape changes on each real. The standard tone is 3.15 kHz +/- 0.2% (or commonly +/- 60 Hz). Another tone would work, but I don't know what the +/- tolernace would be.

If you have a home/project studio and do everything in house it doesn't matter so much. But if you ever want to collaborate or have something outsourced it's still best to maintain industry standards, especially with your mastering deck.

One could indeed rig something with sensors and metal sensing tape spaced 15 inches apart (or whatever speed you need) to get a “Close enough for rock ‘n roll” measurement and I’m sure ZEE is someone who could do it. IMO, it would still be a lot more cumbersome than reading a frequency on a meter while turning a trim pot. And since everyone really needs at least a basic calibration tape anyway I think it would be best to choose another tone… maybe 2 kHz or 4 kHz.

Ethans idea would work as well. One could use leader timing tape and a stop watch.

I knda like miroslav's radar gun idea too.

People can drive themselves batty trying to figure out how to make a speed tape with just a tone generator because it will never work with just that alone.

If you record a tone at a given speed and play it back at the same speed the tone will always be correct even though the speed may be way off from standard spec. That’s why I use the example of trying to build a scale with no known reference weight. You may achieve balance but you’ll never know anything more than the thing you’re weighing matches the weight of the rock you found… whatever that is.

 

[Dr ZEE]

If you record a tone at a given speed and play it back at the same speed the tone will always be correct even though the speed may be way off from standard spec. That’s why I use the example of trying to build a scale with no known reference weight.

Tim, the point here is that to calibrate tape speed you DO NOT NEED to have a special "reference" tone that (as we MUST trust, of course ) is super accurate and was recorded at super accurate and super stable speed.
But! you do need a good (stable) tone (or puls) generator and an accurate counter. And, yes, the machine you record the tone/puls on should have 'acceptable' speed stability.
The actual frequency of the tone and actual speed that it was recorded at are not in the "menue" here.
If you have stable puls/tone generator and the machine's speed is acceptably stable then you record what ever frequency tone/puls on tape first. Second you "measure" the actual "physical" period (distance between pulses) on tape (that is theoretical point), but practically (as per Ethan's method) you cut a specific length of tape and use "accurate" counter to count the total amount of pulses on that length of tape. This is the same (theoretically, as you can calculate it) as knowing what the distance between pulses on tape is (the distance between periodic marks, that is).
The rest is math.
Once you know the physical distance between the marks (pulses) or total amount of periodic pulses in a measured length - YOU CAN use this tape to adjust the tape speed to WHAT EVER speed you want, by, first, calculating "output signal frequency for the desired speed" and using frequency counter to match that frequency while adjusting speed.
In fact, if you know period (distance between periodic pulses) that are recorded on tape, the frequency counter acts like a Speedometer.

I'm not sure why I'm getting into this ... heh heh
Sure this whole idea and especially practical side of it is not for everyone. Yep, it's for lab-heads only. More power to them
***********

BTW, MRL is a real great "entity", but it's also a business The things that MRL sells are not gifts from God, you know, those things were actually "made". But, then, of course, only guys at MRL with their out of this world machines and with their knowledge and wisdom can do this. Of course!!!!



/regards

 

[miroslav]

I wasn't trying to make fun of Ethan's quest for perfect speed calibration...I was just throwing a little fun into the discussion.
Being somewhat anal about accuracy and precision myself, if I knew my deck was running at 14.97ips and I could make it go exactly 15ips without too much fuss...well yeah, I would probably want to do it.

Not sure if measuring out a 25’ section of tape and marking it up with steady pulses would be a simpler/more accurate way than using the MRL Speed tape. Plus…wouldn’t you need a bit more than 25’ of tape? At 15ips…that’s only like 20 seconds. I would think for better accuracy/stability you would want a much longer test tape, like at least a couple of minutes worth or more...yes?
But in the end (at least for me) it’s about using the easiest/fastest/best method to get it done when dealing with rudimentary things like calibrations/alignments. Sure…one can indulge the imagination with other theories/methods, but in the end it will be academic, so for my time…using an already prescribed method makes the cost of the MRL tape less “painful” than any other approach…and then I have more time to play my guitar.

My Otari Alignment Procedure manual for my 5050BIII 2-track describes the process quick and straightforward, though yeah, you would use the MRL Speed tape….and it has an allowable +/- 20Hz deviation from the 6300Hz expected frequency…so it’s not going to be perfect….plus there’s the tape stretch and tension thing to add into the equations.
I’m wondering if they use 3150Hz (7.5ips) and 6300Hz (15ips) for a specific reason rather than say, 1000Hz or some other frequency…?...or was that just what they picked as the standard?

Of course, MRL suggests that the most accurate way is to use a large pulley having an accurately-known diameter…:
” Some transports use such a pulley for tape playing- time measurements. For example, the Studer A-80, which we use to record these Flutter and Speed Tapes, uses a pulley whose circumference is exactly 190.5 mm (7.50 inches). The output of the sensor on that pulley is brought out to the back-panel Remote Control Connector (pin 36 = Return, pin 34 = "Clock"), as a 1 Hz square wave, 0 or 12 volts. We connect this through a 10:1 frequency divider to a period counter. Thus we can make a very accurate tape-speed measurement—one count in a 10.000 s period measurement is 0.01%.”

Luckily, my 16-track deck is completely microprocessor controlled and has built-in SMPTE...so the deck self-adjusts as needed using its internal clocking mechanism. I guess if you hooked up an external sync box to your deck (assuming your deck has that capability), you would not need to worry about tape speed, as the sync box would adjust the capstan servo on-the-fly.

Of course - However MRL says that the alignment tapes are not set up for speed adjustment.

Are you referring to their “basic” alignment tapes?
The link I provided at the top of the thread was to their "3150-HZ Flutter and Speed Test Tapes"...which is how it's described by MRL...and they explain in the PDF how to do the speed measurement.

 

[miroslav]

If you have stable puls/tone generator and the machine's speed is acceptably stable...

Mmmmmmmm...but isn't that like the chicken-n-egg question...?

You can't make an accurate speed tape if you don't know the accuracy of your deck...and you can't know/correct the accuracy of your deck without an accurate speed tape.

Which brings us back to MRL....right?

 

[Dr ZEE]

Mmmmmmmm...but isn't that like the chicken-n-egg question...?

You can't make an accurate speed tape if you don't know the accuracy of your deck...and you can't know/correct the accuracy of your deck without an accurate speed tape.

Which brings us back to MRL....right?

"...machine's speed is acceptably stable"
Speed stability and speed are not the same things.
So, nop, no chicks and no eggs nowhere in tha'house...

/later

 

[miroslav]

I agree...but I'm asking, how would you know if you have acceptable stablity.

Wouldn't the MRL Flutter, Wow & Speed tape be the best way to make that conclusion?
If you have poor stabilty, will it not affect overall speed?

 

[evm1024]

I was saying that my"MRL" (not speed) tapes specifically say that the tones are not of speed calibration quality. Buy the speed/flutter tape if you want to do speed cal.

There may be other tapes that do both.

The only reason to make your own speed calibration tape is to save the $100 (and to have fun).

I thought about using more than 25' of tape. The main advantage is more time to adjust your decks speed. But then again 20 seconds is a good time for that.

The key in the tape length is getting a precisely known tape length. Longer tapes have more handling difficulties. I can measure 25' with good accuracy but might have significant errors with 250'. What we are looking for is known number of cycles in a known length. Which gives us the physical wavelength and allows us to calculate the playback frequency at any given speed.

We can count cycles with great accuracy. We can get length with reasonable precision. The rest is math.

Regards, ethan

 

[miroslav]

I was saying that my"MRL" (not speed) tapes specifically say that the tones are not of speed calibration quality.

Got it.

The only reason to make your own speed calibration tape is to save the $100 (and to have fun).

One man's pleasure....

No, seriously, I can understand about getting into a "gee whiz" frame of mind...we all have something that tickles our curiosity like that.

Would love to hear how you make out if you do decide to try it.

 

[evm1024]

I agree...but I'm asking, how would you know if you have acceptable stablity.

Wouldn't the MRL Flutter, Wow & Speed tape be the best way to make that conclusion?
If you have poor stabilty, will it not affect overall speed?

That's easy.

record a tone (any tone but let's say 1 KHz) on 2 channels. Rewind and play that tone back listen to the tone on one channel from the sync head and the other from the playback head. Listen to the beat frequency. Or send it to a scope in XY mode.

If the beat frequency is 0 you have complete speed stability. If the beat freq changes then you have speed changes.

This works because on record we put the same frequency to tape on both tracks and on playback we compare the tones that were recorded at different times and thus different speeds.

The lissajous figures are quite sensitive to phase differences caused by differential speed variations between the 2 places on tape.

If the figures open and close like normal tape tension issues then you have great speed stability. If the figures are dancing around like a brake dancer then you know that you have significant speed problems.



--Ethan

 

[miroslav]

Rewind and play that tone back listen to the tone on one channel from the sync head and the other from the playback head. Listen to the beat frequency. Or send it to a scope in XY mode.

If the beat frequency is 0 you have complete speed stability. If the beat freq changes then you have speed changes.

Mmmmmm....wouldn't the beat frequency change "equally" at both heads even if there is a speed change?

Do you mean listen to the playback against the source?

 

[evm1024]

Mmmmmm....wouldn't the beat frequency change "equally" at both heads even if there is a speed change?

Do you mean listen to the playback against the source?

You only use the source once, to record. Playback uses different heads (sync for one channel and playback for the other) which take the original signal and compares them with a time offset.

here is how they are recorded;

11121222322211234322221111 Left sync head
11121222322211234322221111 Right sync head

The numbers represent speeds and both tracks are in sync.


And on playback we get:

11121222322211234322221111 Left sync head
-----11121222322211234322221111 Right playback head (offset in time)

Now we see differences in the frequency between the channels.

--Ethan


OK re-edited to get the offset - the bbs took out my spaces......