Messages

1

Tape Tech / Re: Oscilloscope for RTR Alignment.

« on: September 01, 2009, 12:49:40 AM »

I worked on those big tape drives briefly. There is no way they could be economical enough for use in anything but a mainframe. They were very complex. The vacuum columns were not for smooth drive, they were to permit reading and writing tape record by record. They could start and stop in a millisecond. The vacuum capstans had voice coil valves behing them to switch between air and vacuum. The columns had opto sensors to control the reel motors, which would push tape in and pull it out. Flutter was not a big issue.

2

Tape Tech / Re: Oscilloscope for RTR Alignment.

« on: August 31, 2009, 12:19:10 PM »

Keith Johnson first turned me on to the capstan refinishing thing. There was a shop in the Bay Area that resurfaced the capstan motors for the Ampex AG350's. It was some sort of bead blast finish. Rule 1 on AG350s, use the Bodine motors. The Ashland motors had design problems that limited their performance. Something about fields or heat, I don't remember what.

Keith also didn't like the stepped capstan. When I built up a 3M M23 for Reference Recordings (25 years ago now) he insisted that the pinch rollers be replaced with flat ones to defeat the differential drive. I'm not going to argue even though I'm not convinced. I would love to build a Steven's clone and ditch the pinch rollers completely. Or a machine like this: http://www.labguysworld.com/ChesterNewell_2.htm with no loop. But I have a real world to deal with.

If I get the 3M running well I'll invest in Manquen's servo upgrade for the drive.

3

Tape Tech / Re: Oscilloscope for RTR Alignment.

« on: August 30, 2009, 09:34:09 PM »

I should have a signature below. I use the 1audio in many audio boards for consistency.

I just tried a close in measurement of a 1 KHz oscillator with an FFT running at 131K points and 96 KHz sampling. 10 averages didnt take long (but averaging might remove the flutter artifacts??) The frequenct resolution is there, I can see an artifact from something 20 Hz from the 1 KHz tone at -80 dB. It brings up lots of questions I have about digital systems since it should not be there. I'll dig out the analog spectrum analyzer and see if I can find it in the source (an analog oscillator). However its not easy to see from this if its a frequency modulation of the carrier or a tone from something else. I'll dig out the Mincom and see if it sees anything.

If you have lots of time you could look at the "violin string" resonance of the unsupported tape with an optical system. You might learn something. Fixing it would be the challenge. Dale Manquen suggests that above 12 KHz the tape itself damps the vibrations but below a lot can happen.

I have had to put in my own calibration resources. the ST1510 is a real pain to work on but seems to be very stable once its running. The Mincom also is very stable, self calibrating and seems to be built to run forever. Mine is at least 30 years old. Its an IRIG version with a lot of different carrier frequencies. But only the first few are useful.

The 3M design was based on stretching the tape with a stepped capstan and pinch rollers. It seemed to work very well. Some hate the design on philosophical grounds. I hope to have mine up by CES and I can get a better fix on it by then.

There was a process for resurfacing capstans with a satin finish that improved the grip and was supposed to reduce the flutter. That may all be lost art today.

4

Tape Tech / Re: Oscilloscope for RTR Alignment.

« on: August 28, 2009, 11:01:45 PM »

I have an ST1510 and the closest to an Altair in the broader market a Mincom wideband flutter analyzer. For working on tape machines these have lots of benefits over a PC program with the special functions for optimizing bias and tape motion analysis. You can analyze wideband flutter with an fft analyzer but its hard. To get down to 1/2 Hz you need a very large sample and plenty of processing time, and you are looking at close in sidebands of the carrier. The dedicated machinery gets the info much more quickly.

For a scope for this work and one that will be harder to inadvertently damage I would suggest a Tek 5000 series scope from eBay like this http://cgi.ebay.com/Tektronix-5115-Storage-Oscilloscope_W0QQitemZ330354826888QQcmdZViewItemQQptZBI_Oscilloscopes?hash=item4ceaae9e88&_trksid=p3286.c0.m14#ht_500wt_1166 . They can be had for peanuts and the plug in options are endless and cheaper still. Digital scopes have appeal BUT your into analog recording aren't you (or you would not be here) and analog scopes are much better for seeing what comes out of a tape machine that a digital scope. Tape isn't really stable and a digital scope will average the signal making it harder to see.