DIY Capstan Speed Control

Recording Techniques, People Skills, Gear, Recording Spaces, Computers, and DIY

Moderators: drumsound, tomb

Post Reply
thechrisl
audio school graduate
Posts: 12
Joined: Wed Mar 07, 2012 3:03 pm

DIY Capstan Speed Control

Post by thechrisl » Wed Mar 07, 2012 5:18 pm

Hi all, first post here.

I have an Otari MX-50 and would like to experiment more with vari-speed effects. The built in vari-speed is pretty mild unfortunately (less than +- 1 step in pitch) but there is an option for external capstan control via the 37 pin remote connector.

The manual indicates it can accept a range of 4.8k-19.2kHz with a duty cycle of 40%-60%. I'm not sure about the amplitude. That's -50% to +100%, which would be great.

I'm thinking I could get an oscillator (hardware or software based) to generate a square wave and act as a pitch control for the Otari. Has anyone ever tried anything like this (not necessarily on this particular model)? Is this all crazy talk?

themagicmanmdt
george martin
Posts: 1347
Joined: Mon Nov 13, 2006 8:47 pm
Location: home on the range

Post by themagicmanmdt » Wed Mar 07, 2012 8:12 pm

wow - seems interesting that the otari is looking just for a sine generator??? cool!


the old school varispeed DIY is totally well documented at the Ampex discussion list (I found it there originally, to my surprise, they definitely were old dogs rolling their eyes that I was the 394304th person to ask about it, apparently...) - but here's the jist of the 'whole shebang':

other capstans can be fed a 60hz (think mains frequency) sine wave through a power amplifier (like a Crown, or something) that can amplify it to ~120v sine wave (since we run on 120v AC 60hz power...). then, by changing the frequency on the signal generator, we change the speed of the capstan. bam, varispeed. it has it's limits, otherwise the speed won't be reliable, we'd burn it out, or burn the crown amp out.


however, that's mainly for older machines. different machines power capstans differently, and it seems your otari is just looking for a sig gen?

i say, you should totally have a sig gen ANYWAYS, so grab one and try it!
we are the village green
preservation society
god bless +6 tape
valves and serviceability

*chief tech and R&D shaman at shadow hills industries*

thechrisl
audio school graduate
Posts: 12
Joined: Wed Mar 07, 2012 3:03 pm

Post by thechrisl » Thu Mar 08, 2012 6:54 pm

I assumed since Duty Cycle was mentioned, it would be a square wave, not a sine. I read some about what you referred to (using mains frequency) and I think it's the same general idea, but with a twist or two...

Here's the page from the manual. It's all I really have to go on. What isn't clear is amplitude of the control signal, type of voltage on pin 23 (AC or DC?) and the grounding scheme overall. I'm a bit worried about burning something out if I get carried away experimenting.

Thanks for the response, by the way.

Image

Nate Dort
tinnitus
Posts: 1039
Joined: Mon Mar 08, 2004 6:07 pm
Location: Detroit
Contact:

Post by Nate Dort » Fri Mar 09, 2012 6:05 am

thechrisl wrote:I assumed since Duty Cycle was mentioned, it would be a square wave, not a sine. I read some about what you referred to (using mains frequency) and I think it's the same general idea, but with a twist or two...

Here's the page from the manual. It's all I really have to go on. What isn't clear is amplitude of the control signal, type of voltage on pin 23 (AC or DC?) and the grounding scheme overall. I'm a bit worried about burning something out if I get carried away experimenting.

Thanks for the response, by the way.

Image
I have an MX-50, but I've never explored this option. However, I can offer some insight based on what is shown on that page.

First, the amplitude of the square wave for pin 20 is given by the the "Input Signals" note (*6). It's a standard logic low/high of 0 to 5 volts. As you said, the frequency can be between 4.8 to 19.2 kHz, with 9.6 kHz being the nominal speed for whatever mode you're in (3.75, 7.5, 15 ips). This would be really easy to generate via freeware.

Pin 23 enables the external clock to control the speed. The column marked "level" tells you what kind of signal it wants to see. In this case, it wants a logic "low", so you'd just ground pin 23 to enable the external clocking. You could put it on a switch if you wanted, or just hard-wire it on the connector. No need for external voltage (DC or AC). Just tie it to pin 36 or 37.

thechrisl
audio school graduate
Posts: 12
Joined: Wed Mar 07, 2012 3:03 pm

Post by thechrisl » Sat Mar 10, 2012 11:05 am

Here's one more clue from the manual (for posterity or whatever -- you probably have this already):

When the Speed Mode is set to EXT, the tape speed is controlled by the reference signal connected to the REMOTE connector on the rear panel. If the Speed Mode is set to EXT and the PITCH ENABLE line (pin 23) is not activated (pulled low), the FIX indicator will will remain lit and the EXT indicator will flash.

So what you said made sense, by grounding pin 23, it should be pulled low and enable external control. I need to get a cable (or at least rig up a couple pins) but will give this a shot. Something like AudioMulch should work for the signal. Thanks again!

wkrbee
takin' a dinner break
Posts: 199
Joined: Sun Jul 09, 2006 11:06 am
Location: Los Alamitos CA

Post by wkrbee » Mon Mar 12, 2012 8:40 am

Lots of pro tape machines used freq.control of the capstan moter.It was used when the tape machine was locked to SMPTE and needed varispeed to stay locked to the Smpte source,(square wave),usually another tape deck or video deck.The enable pin says low.That means"active low" which means the function is true when the pin is forced to ground,(usually with a pull -up resistor).
Boston uses a MTR 90 capstan motor to pitch-bend their Hammond B-3,but has broken the clutch assy multiple times.
Grandpa sez"common sense ain't that common"

thechrisl
audio school graduate
Posts: 12
Joined: Wed Mar 07, 2012 3:03 pm

Post by thechrisl » Mon Mar 12, 2012 3:41 pm

I'm waiting for a 37 pin connector w/ solder points so I can make secure connections.

Far as I can tell (from a very thorough manual!), pin 23 has a 10k resistor between it and the power supply. So I think/hope I'm covered on the pull-down.

I realized getting a clean 19kHz signal is harder than I thought. At 44.1, most software oscillators will have some aliasing up in that range. Maybe a big deal, maybe not. For now my trusty Pro-One resonant filter is providing a fairly clean signal that can be adjusted with the Cutoff. If I get deeper into this I may get a standalone oscillator (which can be a bit pricey if you want fairly accurate control and a frequency counter so you know where you are). For example, the Rolls MO2020 has one little knob that goes from 20-24k.

But I'm getting ahead of myself... If all goes well I'll try & post a sample of, say, a fake Nashville tuned vari-speed guitar or something like that.

thechrisl
audio school graduate
Posts: 12
Joined: Wed Mar 07, 2012 3:03 pm

Post by thechrisl » Wed Apr 11, 2012 7:35 pm

Well it worked, finally. Nate was right about the standard logic thing. The symmetrical square wave (via synth, soundforge, whatever) didn't cut it. I needed a positive only signal & ended up getting a used function generator that could spit out TTL.

But after all that it works great. I get like plus or minus one octave. I know I can do all this with a sampler but it's just not the same.

So anyway, here's a little track with some slowed down drums. The acoustic gtr on the right side was recorded after speeding the tape up by a 4th, then dropping it back to normal so it sounds like a baritone. Thanks again for the help!

http://soundcloud.com/perol-eyas/soda-p-p

Post Reply

Who is online

Users browsing this forum: No registered users and 18 guests