RCA BC-6B Console Project

[systemtruck]

Gotcha. Understood! That’s funny that you say that about other RCA boards because every one I’ve looked at the schematic of has had this line booster amp. Can i ask which models don’t have it?

I was going to again mention the theory that perhaps all of this extra gain, and then voltage dividing, is solely to achieve the mono/dual switching function… like I mentioned before, so that two channel bussing loss was met with a similar loss during dual mode, just for user friendliness, when they already needed the summing resistors…

BUT… That theory is now negated because the mono BC-5B and BC-3B boards both also have the line booster amps, as attached below.
And in their post voltage dividers, their values are 18k/6200, which is only a slight difference from the 18k/8200 from BC-6A above.

So even thought they don’t need switching or summing, they kept a significant voltage division/loss between the Booster and PGRM.

So what is the deal then? RCA would not have wasted all that expense and parts for no reason. Doesn’t there have to be SOME kind of operational / sonic benefit to have a summing amplifier be separate from the PGRM output amplifier? Like maybe the 4 stages of 12AY7 is preferable for whatever reason? And the two stages of 12AX7 amplification is a better setup for the Push Pull that follows? I’m just stabbing in the dark here.

And what about that pair of 100K resistors that feed the PGRM 12AX7 inputs in the actual PGRM schematic a few posts ago? How does that relate to RCA’s overkill gain staging?

 

[ruffrecords]

Many years ago, when I was first getting into designing tube mic pres and mixers, I purchased CD containing many of the RCA tube mixer manuals so I could study their circuits.
My recollection is that most of them prior to the 6B did not include boost amplifiers. I am not sure if I have the CD or even a PC that can load it so at the moment I cannot give a concrete example. However, I just searched the forum and to my surprice I see the BC-3A uses booster amps to so maybe mt recollection was wrong.

However, it seems to me that you are designing from scratch using the 6B amplifiers as building blocks, which is an excellent idea but, instead of working backwards from what RCA did it might be more instructive to look at what you need to achieve.

Cheers

Ian

 

[systemtruck]

Many years ago, when I was first getting into designing tube mic pres and mixers, I purchased CD containing many of the RCA tube mixer manuals so I could study their circuits.
My recollection is that most of them prior to the 6B did not include boost amplifiers. I am not sure if I have the CD or even a PC that can load it so at the moment I cannot give a concrete example. However, I just searched the forum and to my surprice I see the BC-3A uses booster amps to so maybe mt recollection was wrong.

However, it seems to me that you are designing from scratch using the 6B amplifiers as building blocks, which is an excellent idea but, instead of working backwards from what RCA did it might be more instructive to look at what you need to achieve.

Cheers

Ian

It’s true! I have a pretty settled plan in place for the overalls. It’ll include some basic schematics from your pan notes and your summing notes. I thank you for that!
It will be a simple mixer though. It is not really a full console for studio needs like auxes and headphones and speakers. This unit will only be for having a bunch of preamps with direct outs but also have the ability to sum certain channels together for processing during recording. So I really just am building preamps and stereo summing.
So now im just left so curious why RCA would have a doubled up mix amp situation going. I will probably take your advice and eliminate the booster amps. However then I’ll always wonder what would it would have sounded like with them, and with keeping that voltage divider in place too. So I may still build it with the two boosters, and if there really is no benefit I will just repurpose those two boosters into being their own preamps.
Regardless, I’ll try to post a complete block diagram and schematic here soon.

 

[emrr]

Consider the interstage amps here to be both isolation and driver amps. This console skipped preamp output transformers and pgm input transformers. You also can’t put a pair of passive mix busses straight into a mono sum successfully without an iso amp before the final 2->1 mix. Every previous RCA console has full in/out transformers and lo-Z mixing. This is the final cost conscious version before SS took over, and those returned to full transformer complements and lo-Z mixing.

Interstage amps, again- the only true stock manufactured LCR American tube console is the Altec 250SU, with interstage/buffer/iso amps between the 3 busses and the 3->2 mixing stage.

 

[ruffrecords]

Interstage amps, again- the only true LCR American tube console is the Altec 250SU, with interstage/buffer/iso amps between the 3 busses and the 3->2 mixing stage.

What about the Putnam consoles?

Cheers

Ian

 

[emrr]

What about the Putnam consoles?

Cheers

Ian

Those were all custom one-offs. There is no factory Putnam console. Offhand I don’t recall there being much info about them beyond the amplifiers. The later 610 as far as I remember was a modular channel block, I have only seen custom builds using it.

 

[ruffrecords]

Those were all custom one-offs. There is no factory Putnam console. Offhand I don’t recall there being much info about them beyond the amplifiers. The later 610 as far as I remember was a modular channel block, I have only seen custom builds using it.

I think you are splitting hairs.They were American LCR consoles made in the Putnam factory. At Neve we made loads of custom consoles but they were still made in the Neve "factory".
The 610 modules have a LMR switch right at the top. I think I have a block diagram of the console somewhere. I will see if I can find it.

Edit: Long before the 610 there was this Putnam consoles with left, mid and right channels:

https://funkwerkes.com/web/wp-content/techdocs/MixedProAudio/Putnam-Recording-Console.pdf

Cheers

Ian

 

[systemtruck]

I think you are splitting hairs.They were American LCR consoles made in the Putnam factory. At Neve we made loads of custom consoles but they were still made in the Neve "factory".
The 610 modules have a LMR switch right at the top. I think I have a block diagram of the console somewhere. I will see if I can find it.

Edit: Long before the 610 there was this Putnam consoles of left, mid and right channels:

https://funkwerkes.com/web/wp-content/techdocs/MixedProAudio/Putnam-Recording-Console.pdf

Cheers

Ian

This thread is a goldmine.

Funny that the 610 is brought up.. I am already committed to building at least four 610 preamps for this board instead of being all RCA, so maybe half 610’s and half RCA, and in fact may make all eight preamps the 610. I have some output transformers in hand for the 610, as per David from Cinemag’s guidance. Ready to build and test these preamps.

I found some older RCA board schematics, such as this one https://www.benmook.com/tech/documents/Schematics 2/Schematics 3/RCA 76-B2.pdf, and i see, as you guys stated, that the preamps have output transformers and that the program input has an input transformer.

At this point I’d like to follow the Keep It Simple Stupid method and just make the decision based on the goal.. Let’s say i do eight 610 preamps, and those have 600 ohm output transformers. I will have panning and volume and direct outs, and no other routing whatsoever. And learning what I’ve just learned here, I WANT to use an input transformer on each PGRM amp. And skip the “booster” stage.

So what tube PGRM amp should I use? The Putnam? This RCA 76 i just linked? Or should i try to dig up the 610 summing amp schematics? I tried but couldn’t find them, not surprisingly. I think if i had those, i would have my design already.

I think i need to make this decision first, and then i can decide the resistor values of all my panning and summing.

 

[ruffrecords]

Tubes are not as quiet as semiconductors so a tube mic pre needs an input transformer with a reasonably high ratio (5 or 10 to1) to achieve acceptable noise performance. Most people need/want a direct out theses days to a 610 with its transformer balanced output and step up transformer input is a reasonable choice as the basis of an input channel.

Then you need to decide how many buses a channel may be asked to drive. You will have noticed that the 610 Echo Send is taken form the plate of the output tube via an 82K resistor to a 10K pot. There are two reasons fro this. First, because of the step down output transformer. The level on the plate is several dBs higher than the balanced output so the 82K/10K pot divider will reduce it by about 20dB. The second reason is that the plate output impedance is relatively high so it cannot directly drive a low impedance but 90K plus the output transformer is OK for it. Bottom line, if you want to use a pan pot you need to ensure it does not significantly increase the total plate load much beyond the original design. At the same time you do not want to drop the level too much because with say 8 channels the bus level will be 18dB lower still and the more bus gain you need the more noise you will have.

Alternatively you could consider adding a cathode follower buffer fed from the output plate then you could drive heavier loads without loss of level.

Either way, first thing is to decide exactly what you want it to do and draw yourself a block diagram.

Cheers

Ian

 

[emrr]

I think you are splitting hairs.They were American LCR consoles made in the Putnam factory. At Neve we made loads of custom consoles but they were still made in the Neve "factory".
The 610 modules have a LMR switch right at the top. I think I have a block diagram of the console somewhere. I will see if I can find it.

Edit: Long before the 610 there was this Putnam consoles with left, mid and right channels:

https://funkwerkes.com/web/wp-content/techdocs/MixedProAudio/Putnam-Recording-Console.pdf

Cheers

Ian

It’s hardly splitting hairs- there were hardly any Putnam consoles, period. They existed in his studios, period. There was no factory making consoles on a day to day basis like Neve, ever. You’re drinking the kool-aid if you think they were ever any sort of player; that’s a marketing legend pushed by the modern UA. I do not think you or anyone else can produce evidence to the contrary.

I repeat- you could not buy a stock american tube console with LCR mixing outside of the Altec 250SU. They don’t exist outside of custom builds, and I’ve never seen a BRANDED american LCR custom console either. Only in-house builds identified by the studio or record label.

It’s an important distinction contrasting perhaps britain where virtually everything was custom BBC or in-house designs with few off the shelf pro mixers available for purchase, and america with dozens of players selling stock products. This speaks to technique and market demand. After the tube era, plenty of stereo with assignable busses and pan pots; the territory of Neve.

By the time stereo truly cements it’s place, the technology has all changed.

To the point - the Putnam custom console has interstage/buffer/iso amps like the 250SU and the RCA BC-6 to allow for separate bus mixing, unlike everything else in america MADE WITH A STOCK PART NUMBER TO BE SOLD AS A STANDARD PRODUCT which had no combining busses, only single or dual mono capabilities, so no need for interstage amp blocks.

 

[systemtruck]

Tubes are not as quiet as semiconductors so a tube mic pre needs an input transformer with a reasonably high ratio (5 or 10 to1) to achieve acceptable noise performance. Most people need/want a direct out theses days to a 610 with its transformer balanced output and step up transformer input is a reasonable choice as the basis of an input channel.

Then you need to decide how many buses a channel may be asked to drive. You will have noticed that the 610 Echo Send is taken form the plate of the output tube via an 82K resistor to a 10K pot. There are two reasons fro this. First, because of the step down output transformer. The level on the plate is several dBs higher than the balanced output so the 82K/10K pot divider will reduce it by about 20dB. The second reason is that the plate output impedance is relatively high so it cannot directly drive a low impedance but 90K plus the output transformer is OK for it. Bottom line, if you want to use a pan pot you need to ensure it does not significantly increase the total plate load much beyond the original design. At the same time you do not want to drop the level too much because with say 8 channels the bus level will be 18dB lower still and the more bus gain you need the more noise you will have.

Alternatively you could consider adding a cathode follower buffer fed from the output plate then you could drive heavier loads without loss of level.

Either way, first thing is to decide exactly what you want it to do and draw yourself a block diagram.

Cheers

Ian

Thanks for this rundown! Fully understood and much appreciated.

It leaves me with one question..

Now learning about the ‘older’ RCA boards having more transformers instead of that extra booster stage, i did find one clearly illustrated example so far which is the BC-2B, and I’m sure there are more… https://www.worldradiohistory.com/Archive-Catalogs/RCA/RCA-BC-2B-Consolette.pdf
I’d like to implement this, and have the preamp’s output transformers go through bussing and then hit the PGRM input transformer, and the transformers I’m working with already, ie ones i have in hand right now, might actually allow it. But I’d love some confirmation of this before navigating the block diagram..

Firstly, my preamp output transformer, as you know from any 610 schematic, is 600 ohms at the secondary.
That may be too high to hit the kinds of input transformers at the PGRM input in all of these older consoles which have mic-level impedances at their input primary, such as 23 ohms…. however, the mic preamp transformers that i have from cinemag offer a 600 ohms input strapping. (They are 37.5 R, 150 R, 600 R, : 50K) So, if i set it up so that the preamp output 600R goes through bussing and then each PGRM input has 600R at its primary… is that functional? Can passive panning / fading be accomplished between 600R and 600R?

Side note, I’d be fine with a hard switch at the output of each preamp… choose between bussing or direct out. I’ll never use both at the same time, and this may solve some problems with loads and/or levels.

 

[emrr]

By the time you finish with passive bus losses of any significant number of channels you are back down to mic level. The PRE and PGM input transformers are virtually always one and the same.

 

[systemtruck]

By the time you finish with passive bus losses of any significant number of channels you are back down to mic level. The PRE and PGM input transformers are virtually always one and the same.

Yes understood. The transformer into PGRM will then boost it ala mic style, but it will only be at a 1:9 ratio since it is strapped as 600 : 50k.

And I also understand that the Preamp’s output transformer will also reduce level as opposed to pulling straight from the tube plate before it. So we’re talking about some serious dB loss when designing it this way.

But this is how it was done with these boards, right? They took the preamp output signal after the transformer, at low impedance, and sent it through bussing which then hit an input boost transformer at basically the same low impedance.

So it’s impedance matching but it has some passive circuitry between it. I’m going to go ahead and throw together a block diagram but won’t include values of resistors just yet.

I like the idea of transformers being involved in these stages just like they were originally well before solid state began.

 

[systemtruck]

Here’s a rough block diagram with no values in place. The first one is zoomed in to see details better. Second shows the whole 8 channel affair.

This structure seems very close to the older RCA tube console builds, except that there is panning instead of switching between L/R. I’m unsure how to calculate the RCA console impedances but they are very low.

However, to use this “mic pre” transformer at the PGRM input, I’m strapping the primary at 600 ohms so as to bridge from 600 pre-out to 600 PGRM-in (with passive circuits between). So instead of the more common 150 strapping ( which gives a 1:18 ratio with 24dB gain ), it is 19dB of gain at the PGRM input.

But the old RCA boards seem to have a lower impedance at the PGRM transformer input than back at the preamp transformer output. This is a bit beyond me, and I’m wondering if i can in fact set my PGRM transformer to 150 ohms even though my preamp output is 600 ohms, and build a passive network that communicates still.

 

[ruffrecords]

@emrr All I was reacting to was your original statement that "the only true LCR American tube console is the Altec 250SU". It was not my intention to make a big issue out of it.

@sytemtruck Your block diagrams are beginning to clarify things. You need to be careful with the preamp direct out switching because at the moment it leaves the input to the channel fader hanging in mid air. This will at least worsen crosstalk and at worst pick up noise/hum and add it to the mix. It really needs to be grounded when not in use or you could just make sure you turn it right down when switched to direct.

Your mixing is unbalanced and does not have to be at particularly low impedance. There is therefore no reason in principle why you have to have a transformer at the input of the PRGRM AMP. If you do then you will have to slug the bus to match its input impedance which increases bus loss and noise.

Overall, I like the concept because you could mix and match mic pres as long as they are transformer in and out.

Cheers

Ian

 

[emrr]

Some manufacturers used nominal 150 busses, some 600, RCA and some others 250. The differences come out in the wash, don’t overthink it. Actual available parts may dictate best path. Your gain pots may need to be 1K, or they may be fine being 2k5 or 5k for less preamp loading. Can you get that in an audio pot? How does it scale the rest and what does it do to the PGM side?

There were various tricks used to create an overall balance, look closely at the unbalanced mix bus of the BC-2B and note 4 channels hit the PGM from one orientation, and the other 4 are opposite, with a center tap grounded input transformer. Not critical, but kind of elegant thinking. Some busses are grounded, some are not. Some i’ve had better results changing what was original, all of which I’m doing in the absence of close proximity to a 1kW+ transmitter.

Yes, when switching to direct out a similar impedance resistor needs to be subbed into the bus input channel to keep bus gain and noise the same. Otherwise levels will change.

 

[systemtruck]

Thank you !!
I’ve crunched some numbers, and have done my best with it below. I believe i may be missing something regarding gain losses, and I’m unsure about crosstalk completely in this dual pan context.
As for the values I chose.. i copied the relationships between values in the pan section from @ruffrecords dual pan design here, and simply cut them by 1/5th. So as opposed to 47k pan, 10k slug, 50k sum… it’s 10k pan, 2K slug, 10k sum.
Then i actually started with a 1K volume pot, but found that it loaded the preamp output down to around 300 ish ohms at worst.
Then i tried 2.5K volume pot, and it still was only in the mid 400’s.
Then I tried 5K volume pot, and it achieved just above 600 ohms at worst.
Then I slugged the PGRM input transformer with a 600 R resistor, and if i interpret this all correctly, this 600 becomes the bottom half of voltage division reaching the PGRM transformer. So i used that in my calculations of gain losses.

I feel like I’m getting close, but am missing some key ingredients in these calculations.

Lastly, i added a shunt to ground for the volume pot input when direct out is engaged, but am wondering.. should it have a resistor in series? Just wondering about impact on levels in rest of channels when one or several are switched to direct out.

 

[emrr]

Look up Collins 212A or B, then also 212E F or G. 2 different Collins families indicative of the panless approach most companies followed, as addition to the RCA. You’ll find the SS RCA BC-7 or similar, whatever you can find online to also be similar. Want a tube thing similar to the BC-4/5/6 approach, look at the extant early custom tube Neve drawing. I went through a different one for a client and while not the same, was pretty similar to that drawing.

Not able to get into loading math looking at this on a phone. Pan crosstalk largely a matter of quality and routing of ground references.

 

[systemtruck]

Look up Collins 212A or B, then also 212E F or G. 2 different Collins families indicative of the panless approach most companies followed, as addition to the RCA. You’ll find the SS RCA BC-7 or similar, whatever you can find online to also be similar. Want a tube thing similar to the BC-4/5/6 approach, look at the extant early custom tube Neve drawing. I went through a different one for a client and while not the same, was pretty similar to that drawing.

Not able to get into loading math looking at this on a phone. Pan crosstalk largely a matter of quality and routing of ground references.

Gosh I’d love to take a look at those tube Neve schematics but they are very elusive. I’d especially want to see some successful implementation of pan pots. I scoped what i could here on Group DIY but i have to be honest i can’t figure out where to find a lot of things in this platform.
I did find this, which says that you reverse engineered a 10 channel one of those. That is just bonkers ! https://postfade.co.uk/early-rupert...-stories-part-one-1959-1962-the-valve-mixers/

Yeah I’d love to see some panning in this tube / transformer environment.
I also took a look at some of the Collins consoles schematics that i could find, and i see some interesting things. No panning yeah, but the volume situation seems to involve T attenuators or something similar which is nice. I would actually really enjoy using attenuator networks for each channel volume, but that would end up at least a few hundred bucks for just 8 channels of volume. The Collins also seems to have balanced connections from pre-out transformer to PGRM in transformer. But the wiring is tough to follow and I ran out of time. That is pretty appealing for noise control since it would cancel bus noise which would be pretty wow. But that sounds super complicated to attempt to implement with volume AND panning.

 

[emrr]

Interesting Neve recap. Will have to re-read for detail. Hadn’t seen that. The other Neve drawings are posted here somewhere.