My Location Recording Mixer

[Bill Wilson]

This is the mike preamp of the tube mixer I designed in 64, for my on location master taping business. The mixer had 6 mike preamps
that fed a L,R, center switch on each channel. The output of the switches fed the high impedance mix buses; these were fed to the stereo output amps & fed to the top of the 100k record level pot on my Ampex 351-2 recorder.

The frequency response was limited by the UTC transformers. Freq. response= -1db @30cps., flat from 50 to 18kcps., - .5db 20kcps. for those who do not recognize the cps. term, it is cycles per second & predates the hertz designation; (personally I prefer it).

This is a class A circuit as no feedback is used. It is very clean w distortion so low it cannot be heard. In case you are wondering why there is no bypass capacitor across the cathode resistor; that was a trick employed in the early days of vacuum tube design in order to get more gain from a circuit. By not using a bypass cap the gain is less & results in a very quiet mike pre. Eve Anna Manley agreed with this on diy audio, 4-5 yrs. ago.

More to follow.  P. S. In order to see the entire image right click your mouse & select VIEW IMAGE; this will show the entire drawing.
                        Sorry I do not know how to make the image smaller.

 

[Bill Wilson]

The filament connections on the previous post is incorrect. Pins 4&5 are shorted together & 6.3 vac is connected to the shorted pins (4&5) & pin 9. Yes I said AC on the filaments. In the separate power supply there is a 10 watt wire wound pot wired across the 6.3v winding; its value is 100 ohm to 150 ohm. If there is a center tap on the filament winding you tape it back & don;t connect it. The filamets are wired in the mixer chassis  from one tube socket to the next w 20ga. wire of different colors. The wires are twisted & laid flat against the chassis body. When you have determined all circuitry is passing signal properly; you tune out any hum from the power supply using the 10 watt pot. This works perfectly, it is how we built em in the old days.

Today you would want to use metal film 1/2 watt resistors; back then I had to use 1 watt 5% carbon composition resistors in the preamps & for the stage that the mix buses fed.  The tubes were the best grade Mullards, very quiet & non microphonic. In order to insure the pre's would not be overloaded; I took a U47 (my hottest mike signalwise) & screamed into it at a distance of 8", while observing the output of the preamp off the plate with a borrowed O scope. This is how I arrived at using a 3.3k cathode resistor.
More to follow.

 

[Bill Wilson]

The mixer chassis was made of 18ga. steel w a 7.5" x 19" aluminum rack panel for the control surface. The attached drawing is the original I built from. It shows one 12ax7/7025 used for two mike pres. It also shows how the terminal boards were laid out; there was one on either side of each of 3 tube sockets. The mixer was extremely clean & very quiet. I will try to attach a file of some recordings made thru it.

Here is a list of mikes i used on recording sessions: 2ea. Telefunken badged U47's, 2ea. Sony C37's, an Electrovoice 640something broadcast cardioid, 4-5 AKG D19E dynamics (great multi purpose mike), Altec M21, RCA 44BX & BK5B ribbons. I never had any trouble getting enough gain on any of the mikes.

The upper terminal board has a 68k resistor hanging in the air It also connects to B+.

 

[Bill Wilson]

Here is the layout of the 3 channel mix bus. The Switchcraft 130312L lever switch is listed as a positive ON in one direction & Momentary in the other. I found that I could take a pair of long nose pliers & bend the blade on the momentary side & make it a positive ON. This allowed me to use a switch that cost $4.00ea. in 64. Today the same switch costs $35.00 ea. at Mouser. The volume controls were Ohmite hot moulded 100k, log taper pots.

The reason for tying each bus to ground was to maintain constant impedance & guarantee high frequency response down 1 db @ 25khz.
With the channel volume control at max on & signal just below clipping preamp output is 10vrms. output from the mix bus will be .9vrms.
There is approximately 25 db in thru the switches & mix bus. Preamp will handle .01vrms Max. at the mike connector.
The average output from an entire channel amplifier Is approximately 1-1.25vrms, to feed the recorder input.

Next will be the rest of the channel amplifiers.

 

[ruffrecords]

Fascinating stuff Bill. I love the empirical method of determining the cathode resistor values. I am not sure about the unbypassed cathode resistor increasing gain; it should reduce the gain. Could you post a link to where Eve Anna Manley agreed with this?

One word of advice. It might be an idea to reduce the size of your pictures. I usually scale mine so they are no more then 1024 pixels across.

Looking forward to the nest installment.

Cheers

ian

 

[Bill Wilson]

Hi Ian: I did not mean that a cathode resistor w/o bypass cap. decreases the gain,rather than increasing it. My statement was that early designers used cathode bypass to increase the gain of the stage; the downside is that noise is also increased. I had posted this observation on DIYAUDIO & Eve Anna Manley added a post agreeing with it as does the Radiotron Designers Handbook 1952 edition.
She further stated that in the Manley product line, they do not use cathode bypass caps. These posts were 4-5 years ago.

This is the last schematic in the series. The input of the channel amp. is connected directly to the mix bus. In my construction I ran 3 lengths of #16 buss wire, it was supported by the resistors from the channel select switches, soldered to them. NOTE: I did most of my recording in two locations in New York City, but never had a hint of RF on the mixer.

The output of the channel amps. are 600 ohms impedance; this was fed directly into the record level on the Ampex 351-2 tube recorder.
The power supply was in a separate small steel cabinet from Bud Mfg. & was connected to the mixer chassis via a 6' cable.
These circuits are super clean & very quiet, I never heard tube hiss or hum in the headphones. No provision was made for vu metering as I used the 4" meters on the recorder.
I will post sketches of the front panel layout & chassis layout soon.

 

[ruffrecords]

Hi Ian: I did not mean that a cathode resistor w/o bypass cap. decreases the gain,rather than increasing it. My statement was that early designers used cathode bypass to increase the gain of the stage; the downside is that noise is also increased. I had posted this observation on DIYAUDIO & Eve Anna Manley added a post agreeing with it as does the Radiotron Designers Handbook 1952 edition.
She further stated that in the Manley product line, they do not use cathode bypass caps. These posts were 4-5 years ago.

Thanks Bill. That makes a lot more sense.

Cheers

Ian

 

[lassoharp]

I took a U47 (my hottest mike signalwise) & screamed into it at a distance of 8", while observing the output of the preamp off the plate with a borrowed O scope. This is how I arrived at using a 3.3k cathode resistor.

I like this.  It's real world empirical testing and probably had the final say over what was figured on paper.

Here is a list of mikes i used on recording sessions: 2ea. Telefunken badged U47's, 2ea. Sony C37's, an Electrovoice 640something broadcast cardioid, 4-5 AKG D19E dynamics (great multi purpose mike), Altec M21, RCA 44BX & BK5B ribbons. I never had any trouble getting enough gain on any of the mikes.

The tubes were the best grade Mullards, very quiet & non microphonic.

Just out of curiosity, would you comment on why you went with the A-10 vs other available inputs at the time?  Namely HA-100, LS-10, Langevin 400 series, WE 618, Peerless S/20/20 series.  I am assuming that size was a factor for a 6 input portable unit so LS-10 may have created an issue there.  For the others mentioned was it a matter of max level handling vs cost/performance?  Or were the transformers not even given that much consideration compared to the mics and tube choices?  Again just wanting to hear the thought's of a DIYer from a time when a wide range of parts were readily available.(and not subject to the madness that exists today).

I would also like to hear your comments on why you chose to build your own mixer vs using one of the commercially available units at the time.  Was it a matter of some needed customization?  Was building this mixer substantially less expensive than commercial ones new or used?

Thanks for sharing your schematics.

 

[Bill Wilson]

Hi Lassoharp: Glad to answer your inquiry. I was the sole repair/installation tech. for Central Jersey Sound Center, Eatontown, N.J. We were an Altec Engineering Contractor & a custom Hi Fi store; just the two owners & myself. Had been married 6 yrs. & not a lot of money.

I had worked for the largest analog computer manufacturer in the world; all vacuum tubes. Had a buddy in from munich that I worked with in the engineering lab. Wanted a U47 in the worst way, Gerhard said no problem my fiance is coming over shortly. At that time Telefunken was doing distribution for Neumann. You could go to a Tele. radio store in Germany & buy it over the counter. She bought one 47 mike bottle for me & one for my church (I was the engineer for the weekly live broadcast from the church, also the announcer).
The 2 mike bodies $125- apiece & 2 shock absorber mike stand adapters w 30' cable $50- each. I built my own power supplies. Won't go into details how they got in the country; Gotham Audio had exclusive U.S. sales & distribution rights. We were located on the Jersey shore; only 55 mins. to the Lincoln tunnel. Drove into N.Y. & went to G.A. & bought the teuchel chassis connectors form the V.P., very nice guy as opposed to Steve Temmer.

Choice of xfmrs., could have bought peerless at dealer cost; did not want to tip my hand to the guys I worked for. The UTC A10 -1db 30hz., flat 50-18khz., -.5db 20khz., handles +15dbm before core saturation, less room, weight & cost than HA series & LS series not sensible choice. Purchased from local parts wholesaler two at a time, cost $13.20 ea. They were just as good as peerless but not so popular on west coast. The W.E. transformers were no shining star on the high end. As you suspected the only alteration to the mike pre. was the cathode resistor value. I wanted a 3 track Ampex so badly I could taste it; could not afford, so I settled for 351-2, 2 track.

Built the mixer as a 3 channel out mixer, only 2 channel were made commercially. Had a friend who owned a metal fab. shop, where I made my own chassis after work; again limited funds. The co. I worked for were Crown & AKG dealers, paid $55- ea. for 4 D 19E mikes.
My co. name was Audio Engineering Associates circa 1958. I sent Wes Dooley an email; he said I pre dated him on the use of the name by 5-6 years. Thinking of using it again fr my custom MC phono preamps & headphone amps.

 

[lassoharp]

Thanks Bill,  Great story on the Neumann mics.  I can imagine that was a super $$$ proposition trying to get the 3 track Ampex at that time.  Was your interest in that for getting the lead vocalist on a single track and doing the backing music in stereo?

 

[Dr Gris]

Love these stories, thanks for sharing.

Best regards
Magnus

 

[Bill Wilson]

I greatly respected the work that Bob Fine was doing for Mercury Records "Living Presence" classical recordings, using only 3 Omni mikies to capture symphony orchestras. Yes I would have used the third track for lead vocal if there was one. In the absence of track 3 I modified the switch matrix to split the center channel equally to L/R; for solo vocals. I never recorded R & R & am not a fan of "multitrack mono"; which is what most recordings are today. Not many younger mixers have heard true Stereophonic recordings; just can't make them w a pan pot. My style of recording was to find the loudest passage in a number; set the level to hit 0 VU on it & never ride the gain. there were never any EQ's or compressors used; only in the mastering suite, when the master lacquer was cut.

This was particularly effective with a Black choir that specialized in performing Negro Spirituals. The Pastor was the tenor soloist (had a great trained voice); I recorded them in their church in Cape May, New Jersey. It was a small church w a peaked ceiling. The unique thing was that the ceiling was covered with tin ceiling material, that provided a nice natural reverb. A spaced pair of U47's set to the Omni pattern covered the choir & after testing for proper positioning gave a killer stereo image of the choir.

Miked the soloists w an Altec M21 "Lipstick" cardioid condenser. When the lead vocalist came in it was a killer sound. Put this group on my own custom gospel label. They had ordered 1000 LP's & were paying the costs; I begged them to have it cut in stereo but they could not be convinced to do so. The taping was done in Feb. 1965. You can hear one of the cuts by going to
www.soundcloud.com    and entering Macedonia chior in the search box.

 

[lassoharp]

I begged them to have it cut in stereo but they could not be convinced to do so

Due to extra expense?  Out of curiosity, how much was a stereo pressing vs mono back then?  I feel your frustration of having such a recording shrank to mono.  Were you able to save the master tapes/copies for later dubbing?


I've only recorded a few choirs and small orchestral ensembles but I did like the 3 mic set up vs spaced stereo pair. 

 

[Bill Wilson]

Hi Lassoharp: I used Sonic Recording Products In Hicksville, L.I., N.Y. I paid $.30- $.33 per 12" pressing in 1000 piece quantities.
I used a specialty label printing co. in lower Manhattan for labels & custom cardboard sleeves. The only increased cost to me for stereo was $90- per side Stereo & $50- per side Mono, cutting the master lacquer. All other costs were the same. I used Sid Feldman @ Mastertone Studios in Downtown Manhattan for cutting my masters. There was a $25- Courier fee to get the master lacquers from the studio to the plating plant; that was close to the pressing plant. since the plater delivered stampers to the presser almost daily;there was no charge incurred.

Part of the reasoning for having it pressed in mono was that in 1965 there were still many average households that had not converted to stereo, as yet. Playing a stereo pressing more than a couple times w a mono cartridge would irreparably damage the vinyl.

my choice of pressing plant was guided by their clients. If you remember Command Records in the late 50's to mid 60's was a label aimed at the audiophile community. The recordings of the pop & jazz material were made at Fine Sound in N.Y. Bob Fine was also Chief Engineer for Mercury's "Living Presence" classical catalog. He used different miking techniques for studio recordings. The results of his work was always outstanding. The other link in the chain was their disc mastering system. It consisted of a Westrex 3A stereo head mounted on a Scully  lathe, they stripped off the automation; so far nothing different from anyone else. The Westrex head was driven by two Mc Intosh 200 watt industrial amplifiers. Bob Fine had made his own modifications to the amps & feedback electronics, in addition he had George Piros a very talented disc mastering engineer; that made the difference in discs cut at their facility.

 

[usekgb]

I love the sound of 2-track spaced omni or cardioid.  I used to do quite a bit of it.  From the University of Utah marching band, to church choirs to pipe organs.  Imaging the reverb in a football stadium (when I used to have access to it) excited by an entire marching band!  Spine tingling!

 

[joaquins]

I don't have the time right now to see the whole post, I'll the the time later, it seems interesting... But the not bypassed cathode resistor is in fact adding some negative feedback letting the cathode float with the grid, so when the grid goes up, also the cathode and decrease the gain, as NFB is being applied you could squeeze more gain from the tube with less distortion, but you are having NFB and decreased gain in relation to the same resistor bypassed by a big enough cap, at least from the RC knee and above.

JS

 

[fazer]

I have a bunch of UTC-O16's.  I'm going to play around with you circuits Bill.  They are not the same as an A10 but are not bad with a pad.  I noticed A10's were $13. In a 1963 catalog  and O16's were $10 because of their extra shielding. 

I no somewhere there has to be a recommended zobel circuit some where but never found anything .

Do you have a picture of your mixer?

 

[gridcurrent]

I have a bunch of UTC-O16's.
I no somewhere there has to be a recommended zobel circuit some where but never found anything .

as you are probably aware, the O-16 is similar to the O-1 but with extra shielding and primary CT.
contributor AusTex64 has experience with Ouncers with tube preamps, loaded and un.
he may have a suggestion pertaining to a Zobel.

 

[abbey road d enfer]

This is a class A circuit as no feedback is used.

The notion of class A does not rule out the use of NFB and vice-versa. In fact, there is NFB because of the unbypassed cathode, which increases the input headroom at the cost of loosing some gain and increasing the output impedance, thus reducing the operating level of the subsequent stages. In order to compensate the reduced operating level, gain must be added somewhere, which will also increase the output noise.

It is very clean w distortion so low it cannot be heard. In case you are wondering why there is no bypass capacitor across the cathode resistor; that was a trick employed in the early days of vacuum tube design in order to get more gain from a circuit. By not using a bypass cap the gain is less & results in a very quiet mike pre. Eve Anna Manley agreed with this on diy audio, 4-5 yrs. ago.

I would be very cautious with any technical assertion coming from Ms. Manley. IMO she is to an electronics designer what a wine lover is to a wine maker.
An unbypassed cathode can only increase the EIN, first by introducing an additional resistance in series with the equivalent noise resistance of the tube. The formula for equivalent noise resistance is Rneq=2.5/Gm, which accounts for about 1.6kohm. Adding 3.3kohm in series increases this noise term by about 5dB.
Second, unbypassed cathode makes it very sensitive to interference, in particular coming from the heaters voltage.

However, the proof is in the pudding, and I don't doubt this mixer has served you well, with adequate performance considering the limitations of tape as a recording media.

 

[ruffrecords]

I would be very cautious with any technical assertion coming from Ms. Manley. IMO she is to an electronics designer what a wine lover is to a wine maker.

David Manley would have agreed with that assertion but used a lot more expletives in its expression.

Cheers

Ian