Revox A77 restoration

SeleniumFalcon

After the playback head has been aligned with a reference standard and the record head positioned correctly to the playback head the next step is to calibrate the machine to a chosen recording tape. Once upon a time this meant picking a favorite tape from lots of choices. Guys would swear by their favorite whether it be Maxell, TDK, Scotch, BASF or Sony they would have reasons why they picked that particular tape. Just like today's interconnecting cable and speaker wire choices recording tape was an individual matching of components and owner's expectations. Once the brand was selected then there were specific formulations: low noise, higher dynamic range, headroom improvements even playing time lengths were some of the factors. More recent tape decks would have selectors to switch between "regular" and "special" tapes. These factors became even more so as cassettes began to take over the home recording markets.
My choices are a great deal easier since there are very few currently produced tapes available. I decided to try and match the A77 to SM911 tape a modern version of an old classic BASF 911 formula. I also have a reel of unused Scotch 150 tape that was probably a favorite tape when the Revox was around.

SeleniumFalcon

But first, I have a problem to solve. Most of the tape setup instructions involve having the playback level control turned all the way clockwise to the highest setting. Whenever I did that I noticed that there would appear a fairly strong, high frequency signal on both channel I and II that would taint any adjustments I made. I started by measuring the level of the signal, 2.4V on channel II and 0.6V on channel I and a frequency of 145kHz. If I reduced the playback volume slightly everything went to normal. Since I first noticed this problem when I was recording test tones I thought it was related to the recording process. So, I began swaping out the recording boards and the bias oscillator card, but no change. I thought I had a defective volume control so I cleaned it thoroughly and kicked myself for not doing that before I put it in.
But no change, so I decided to swap out the volume control (not an easy task) and in the process managed to damage the plastic frame of the selector switch that goes in front of it. I finally managed to exchange both the volume control and the mating selector switch and hooked everything up and the problem was still there. I next tried to search around for a grounding related cause, using trusty alligator clip leads to try grounding the head covers to the frame, but no change. I suspected the recording bias was somehow contaminating the audio ground since the 120kHz bias frequency was near to the 145kHz that was leaking into the audio signal. More boards exchanged and some slight changes but no real solution. Time to consult the Studer/Revox zen master, Jack Clark. I sent Jack a detailed description and hoped he could find the answer. Jack's responses are always extremely to the point and have no excess information. "Your reproduce trimmers are set too high and the circuit is going into oscillation. First replace R814 on both repro amp cards with a 1k ohm resistor and adjust the repro trimmers to the value in your service manual." Bingo, no frills just the correct answer.

JayCee

SeleniumFalcon wrote: »

Guys would swear by their favorite whether it be Maxell, TDK, Scotch, BASF or Sony they would have reasons why they picked that particular tape.

I can relate. Back in my cassette days I was a TDK SA and SA-X (CrO2) guy. I felt the highs were better. Purchased SA when my funds were low and SA-X when I was rolling in the dough.

SeleniumFalcon wrote: »

Time to consult the Studer/Revox zen master, Jack Clark....Bingo, no frills just the correct answer.

Awesome that you have such a guy. This will be my first public announcement...I recently acquired a pair of Genesis Technologies Gen II speakers, for free. Came about them by way of a very generous gentleman and Paul McGowan...owner of PS Audio and co-designer of the speakers (primary....Arnie Nudell, Infinity fame, RIP, ). There are some accelerometer issues w/2 of the woofers but I have Mr. McGowan's ear and his generous offer to help me get them sorted. Nothing like having a "zen master". In my case, he's THE guy. Speakers are in the basement and waiting as a winter project.

Back to your post...congrats on getting your problem sorted.

SeleniumFalcon

Setting up a particular tape is a two step operation. The first step is to begin recording a high frequency tone (10kHz) at around -20dB and looking at the output with a good quality meter. Then the bias current is increased until a peak is noticed and as the bias strength is increased the meter reading begins falling off. This is called "over-biasing" and is done to reduce distortion during peak recording levels. Depending upon the tape's formulation and recording speed this can be either -4dB or even -6dB of decrease after the peak is reached.
Once the bias has been set the next step is to set the record EQ for each channel and each speed. This is done by first recording 1kHz and then switching the signal generator to 12kHz and note if the level is lower, the same or higher. Adjustment potentiometers are set so the results are the same.
This brought me to the second problem. There was no problem with the Scotch 150, the bias and EQ were adjusted with no difficulty. However, trying to set up the SM911 there wasn't enough EQ adjustment for either 3.34ips or 7.5ips speeds to get the 12kHz the same as the 1kHz. There was 4dB more energy at the higher frequency than the lower one. The SM911 is what is called a "low noise" tape and designed to have very high sensitivity to upper frequencies. Each of the microscopic magnetic particles are made extremely small and needle shaped and packed very densely on the mylar tape. These small needle shaped, high density particles respond to high frequency signals better than their older Scotch relatives. This wouldn't be a problem if the A77 had a way to compensate for this high frequency boost, in fact it would reduce high frequency hiss and record transient waveforms better.
Time to do some thinking.

SeleniumFalcon

JayCee wrote: »

SeleniumFalcon wrote: »

Guys would swear by their favorite whether it be Maxell, TDK, Scotch, BASF or Sony they would have reasons why they picked that particular tape.

I can relate. Back in my cassette days I was a TDK SA and SA-X (CrO2) guy. I felt the highs were better. Purchased SA when my funds were low and SA-X when I was rolling in the dough.

SeleniumFalcon wrote: »

Time to consult the Studer/Revox zen master, Jack Clark....Bingo, no frills just the correct answer.

Awesome that you have such a guy. This will be my first public announcement...I recently acquired a pair of Genesis Technologies Gen II speakers, for free. Came about them by way of a very generous gentleman and Paul McGowan...owner of PS Audio and co-designer of the speakers (primary....Arnie Nudell, Infinity fame, RIP, ). There are some accelerometer issues w/2 of the woofers but I have Mr. McGowan's ear and his generous offer to help me get them sorted. Nothing like having a "zen master". In my case, he's THE guy. Speakers are in the basement and waiting as a winter project.

Back to your post...congrats on getting your problem sorted.

Very cool! I met Arnie "Noodle" many years ago when Infinity was just getting started and a true audiophile and brilliant designer. In fact it was his suggestion that I try my Infinity speakers with some Audio Research gear that got me started on this whole excursion. He also was the first guy to have a Studer deck and some master tape recordings way back in the early '70's. You'll have to document your experiences with the Genesis speakers.

JayCee

I plan to document when the time comes.

Meeting Nudell and getting recommendations...sweet. I'd like to have met him.

For more recent times It's great that many of the big names of audio are so accessible via the net. I've had "conversations" w/Ted Smith, Bascom King, Paul McGowan, of which I own their gear, and a few others in recent years. Also, was cool to meet everyone in Polk HQ and have a BBQ at Al's house during '12 Polk Fest.

SeleniumFalcon

Rick Wakeman and A77.

SeleniumFalcon

An easy way to view the frequency response of recording and playing back is with a real time analyzer (RTA). The first photo shows the response of channel I using Scotch 150 tape. The vertical scale is 2dB.

SeleniumFalcon

This photo is with the deck calibrated for SM911 and using over-bias to get the best response.

SeleniumFalcon

As you can see the response using SM911 is not as linear as when the Scotch tape is used. To try and resolve this issue I first tried increasing the value of the trimming resistor from 20k ohms to 25k ohms. I had looked at the schematics for the more recent Revox B77 and PR99 and both used a 22k ohm variable potentiometer in this position. The results were not any better, in fact it looked like the change made adjustment less responsive. Then I ordered a roll of SM468 tape, understanding that this formulation might be better for an older deck. It was somewhat, but not quite what I was wanting. On a whim I decided to replace the two playback circuit boards with original unmodified boards. A very definite improvement, things looked the way they should. I began to suspect the original calibration tape was the root cause. This tape's high frequency response had caused me to change the values of the feedback resistors to, what I believed, accommodate the characteristics of the new playback head. But if the tape itself was giving me false information then I had made changes that boosted the middle upper frequency response and caused the oscillation and poor recording capability with modern tapes which also increase the upper treble.

SeleniumFalcon

Fortunately I had another calibration tape (165nW/m) and with original value resistors installed in the playbck circuit boards I readjusted the A77 for SM911 and finally got better results (at least on the middle and upper range, the bass bump is something I have to figure out).

SeleniumFalcon

The next part of this project is to build two new playback circuit boards with some (hopefully) improvements. I want to add an emitter follower stage to the playback circuitry. This is using an NPN transistor to provide a high input impedance to the amplification stage before it and a low output impedance to the output. This is similar to a cathode follower stage in vacuum tube designs. Here is a schematic drawing to illustrate what happens.

SeleniumFalcon

Since I need to add a transistor (BC107C) and a 1.0k resistor I needed to add a small circuit board to the original. I removed all of the components on two spare boards attached the two smaller boards and painted them Navy Blue Angel blue (for good luck).

SeleniumFalcon

Progress on this project has been somewhat hampered by a shortage of available parts, namely Dale RN series resistors and WIMA film capacitors. Normally I would place an order with either Digikey, Mouser, Allied Electronics or Newark and within a few days I'd have the parts. Spoiled, I guess. Now Digikey is out of stock on and 330 ohm and 1k resistors and 10.0uF WIMA capacitors. But here's a photo of the two boards with all but one value resistor installed (I changed the color to flat gull grey after scratching the other surface while drilling small holes for the connecting wires).

SeleniumFalcon

In the meantime I started thinking about a portion of the playback circuitry that I might be able to improve. There is a tank circuit (a parallel capacitor and inductor tuned to a certain frequency) that is present to filter out the bias signal that is present in the playback. This is adjusted during setting up by moving a metal slug in a variable inductor while looking at a high frequency capable meter. Here is the schematic drawing and capacitor C808 and variable inductor L801 make up this tank circuit.

SeleniumFalcon

My idea is to first measure the exact frequency of the bias (it's supposed to be 120kHz) and then knowing the exact value of C808 calculate the required value of L801 and use a good quality fixed value inductor instead of the saturation prone variable. As it turned out the bias frequency on this particular machine is 117.6kHz.

SeleniumFalcon

SeleniumFalcon

I remembered I had some small gauge wire and a couple of pot cores for a long ago project, so I decided to try and see if I could make an inductor. I attached the core to my Dremel and used my variable AC to slow it down so I could wind some wire. I wrapped by hand a few layers first then finished with the Dremel setup. I measured the result and was off by a whole magnitude, uH instead of mH.

SeleniumFalcon

SeleniumFalcon

After thinking about this all weekend I decided it was best to completely eliminate the tank circuit instead of trying to build a better one. The circuit is there to filter out the 120kHz bias signal during recording, but for routine tape playback there's no bias signal to need filtering. Since this deck was just going to be used for playing already recorded tapes I could keep the circuit path as simple as possible. But it was interesting to learn a bit about how these filters worked and I like making lots of measurements of small value capacitors and coils.

JayCee

Not sure if you've seen this but a good piece on Mo-Fi. Interesting history of Shawn Britton (Mo-Fi's Senior Mastering Engineer and written by him), his start w/the company, insight on his mentor and the geeky process (gear) you, no doubt, will find interesting.
http://www.enjoythemusic.com/magazine/special_feature/0820/Mobile_Fidelity_Sound_Lab.htm

SeleniumFalcon

Thanks for the link! Good reading.
"a Studer A80 Mark II with a John Curl reproducer amplifier that they called a "SuperMaster" I've heard about this deck, would love to know what he designed.

JayCee

SeleniumFalcon wrote: »

"a Studer A80 Mark II with a John Curl reproducer amplifier that they called a "SuperMaster" I've heard about this deck, would love to know what he designed.

That quote is actually when I thought of posting to you.

SeleniumFalcon

Vangelis with A77 and some big speakers in the background.

SeleniumFalcon

SeleniumFalcon

At last the remainder if the necessary parts arrived and were checked out and installed. The two playback circuit boards are completed, the transistors were sorted for linearity and gain matched. I found four BC109C transistors that were made in Germany by the ITT company and ordered them from Germany. They checked out very nicely and along with various Siemens and others found some good representatives. The emitter follower transistor is in place and connected to the circuit. All of the resistors have been replaced with Dale RN60 versions (several are attached to the bottom of the base dues to space limitations).



All of the tantalum coupling capacitors have been replaced with WIMA film types, Also the 1500uF electrolytic capacitor that is the high pass filter component in the EQ feedback replaced with a low leakage type.This is an important part since any DC that is leaked by this capacitor combined with the 30dB gain of the circuit leads to increased noise in the output. As I began testing these new boards I discovered that one worked fine and the other didn't (the one wired with white wire worked). It took quite a while to find the problem (a poorly installed 3.3uF coupling cap) and get back on track.

SeleniumFalcon

I wanted to be able to power up and test the playback circuit board without the necessity of plugging it into the deck itself. I decided to build a connection platform that would give me access to both sides of the board and be able to attach a DC power supply and connect external volume controls, EQ selection, etc.

SeleniumFalcon

Here I've connected it to my little signal generator, attached my new dual DC power supply and oscilloscope for a trial test.

JayCee

Really interesting interview in the latest edition of PS Audio's Copper Magazine. https://www.psaudio.com/copper/article/martin-theophilus-of-the-museum-of-magnetic-sound-recording-part-one/
Here's an excerpt from Martin Theophilus of The Museum of Magnetic Sound Recording...
"Now [the Museum collection] is hitting 200-plus reel to reel tape recorders. [Also], 100-plus mics, plus mixers and accessories. It important to mention that many tape recorder collections I’ve seen are [simply] lined up on shelves for display. Almost all our [units] are working recorders, connected to sound systems for demonstration, and surrounded by brand-specific ads and accessories. I strive to regularly exercise the machines (not always successfully) and [do] complete maintenance where necessary."

Teaser pic:

SeleniumFalcon

Thank you for posting that link!