Somewhere along the line of product development at 3M, we developed an 8-channel reproduce module. I think the customer was Irv Joel at A&R Recording in New York. The module was the same 5 ¼” height as the rest of our single-channel record/reproduce modules, but it was crammed with 8 reproduce cards and 8 line amplifiers. On the back panel were 8 output transformers, 8 head connectors and 8 XLR3 connectors. In the center of the module was a small VU meter (from Ron Newdoll) and a selector switch that connected the meter to any of the 8 outputs. The power supply was mounted externally on the floor of the recorder’s cabinet. In a way, this was probably the beginning of the product evolution that would eventually lead to the M56 compact 16-track recorder.

With Camarillo less than an hour away from Hollywood, we frequently would visit with customers regarding their requirements. I remember one of those meetings with Bruce Botnick at Sunset Sound. We were talking about 8-track machines since Wally’s 8-track was a frequent resident at Sunset before they bought their own machine. Bruce was discussing the process of overdubbing and he made the comment “I really don’t need a full 8-track recorder. I could get by with just one channel of record if I could move the record channel around. Just give me 1 selectable channel of record and 8 channels of playback and I will be happy.”

After the meeting, I began to think about what it would take to fill his wish. Having only one channel of record would save quite a bit of money and space. We had the 8-channel reproduce module described above, and all we would need would be one record module. The new part would be the head switching for the record side.

Now here we need to discuss one of my greatest attributes as a designer. I was too dumb to know that you couldn’t do certain things. In this case, anyone with a lot of experience might say that you can’t use the same record module on 8 different tracks without adjusting bias and level for each track. My experience was that any track in a given head was usually within 1 dB of the all other tracks as far as bias sensitivity, output level or equalization. I figured that was probably quite acceptable in a real-world situation. This was before the widespread use of Dolby, with the attendant need for close level matching to maintain ‘Dolby level’ alignment.

I talked to Wally about Bruce’s idea of a selective recorder. I convinced Wally that it could be done, and he agreed to order a machine. I would hand-build it at Mincom. There were limitations to the resulting machine, and I made certain that Wally knew them. Most importantly, I emphasized that one channel would be in record, and that all the rest of the channels would be in Sync (Sel Sync to you Ampex folk). This was a result of my planned implementation, which was extremely simple.

All 8 track of the record head would be wired to an 8-position rotary selector switch. Two decks of the switch were shorting wafers, the kind that short out all positions except the desired channel. The rotor is basically a solid disk with a small notch cut out for only one channel. This is usually used to mute unused inputs. I used these wafers to connect all the unused record tracks together, and I fed this composite signal to a Sync repro amp that was identical to the normal sync amp in a record/repro module. The only difference was that this input went directly to the sync amp rather that through the module’s rotary switch for selecting Ready/Safe/Sync.

The switch had 3 more decks, 2 for selecting the record track and 1 for selecting the erase track. The erase heads all had a common ground.

I was able to fit the switch, 16 head cables and 3 output cables into a small Bud box that I mounted on the right-hand side of the transport with the shaft sticking up through the transport’s cover plate near the tape motion pushbuttons.

Everything worked as it should, and the consistency from track to track was quite good. (It would have had even closer matching if we had been using 3M heads in place of the IEM heads, but 3M hadn’t developed any heads at this point in the game.) Wally sent his truck up to get the machine, as he usually did, and off the machine went to Hollywood.

Well, it didn’t take long to hear back from Wally. Next day he was on the phone.
“WWWe ccccan’t uuuse the machine!”
“Why not?”
“AAAll the tttracks are in ssssync!”
“Yes, they are. We discussed that and you said it wouldn’t be a problem.”
“WWWell it is!”

Understandably, the recording engineer wanted to be able to mute certain tracks in the headphones. I didn’t know enough about recording techniques to recognize that this would be a deal killer. I had relied on Wally’s input.

So it was back to the drawing board. The next version of the Selective recorder had two record modules and a control panel with 8 4-position selector switches and 8 toggle switches. Each channel could be selected for Record 1, Record 2, Sync or Off. The toggle switch provided a lockout to avoid overrecording a good track that was to be kept. This provided enough versatility for the intended application, and we did sell a few machines of this configuration.

We even used our prototype 2” transport to make a double version with 16 channels of reproduce and 4 channels of record – just double the hardware. After all, who would ever need to record on all 16 tracks at once! Ha! The one restriction was that the track assignment was split in half, with two modules and one switching panel serving the top 8 tracks, and the other set serving the bottom 8 tracks. We exhibited that machine at the Spring AES Convention (back when there were two a year – Hollywood in the spring and New York in the fall.) Wally promised to buy this machine, but then he backed out when some of his customers said that they wouldn’t rent a partial 16-track. They wanted a fully functional machine. But this starts spilling over into the birth of the M56, which is another story….