Looking for some help identifying and resurrecting this (RCA?) box

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Rothchild

Member
Joined
Aug 21, 2011
Messages
11
Hi there,

A little while I chanced upon this box:

rca%20front.JPG


Emrr was good enough to give me some pointers on GS and after a bit more background work I thought I'd bring it here to see if anyone else would be kind enough to help identify it or to offer further insights, as my Dad and I are trying to resurrect it (he's an electrical engineer of an age where he recognises all the parts in it but he's not really got any experience with audio tech.

The first job he's undertaken for me is to map out the schematic:

http://www.miwi-music.com/MusicAndMedia/rothchild/rca/-RCA-circuit.jpg

And I've got a few more pics for interest and to hopefully jog some recognition from someone. All the backstory I've got is that it came from an RCA PM80 Camera, but to date I've drawn something of a blank on this product (bar one very sketchy pic in a japanese studio).

My Dad has some questions about how some of the specific parts of the circuit work (the de-esser in particular doesn't seem to make sense) and I shall return with these but I just wanted to get this up to see if anyone here can shed any light on it.

http://www.miwi-music.com/MusicAndMedia/rothchild/rca/fullinnersrca.JPG

http://www.miwi-music.com/MusicAndMedia/rothchild/rca/rcavalvestop.JPG

Any help, advice or pointers gratefully received.
Many thanks,
Child
 
Nice drawing there.  The de-ess part is hard to see in this drawing, but I assume is like all the earlier generations in that it increases overall high frequency sensitivity in the side chain.  The result isn't like a modern de-esser, but shelves down the overall top end under compression.  This still appears to be a british made product, and I'm unaware of an RCA US direct equivalent, though the earlier USA versions are all very similar. 
 
As Doug said.
It's hard for me to read all the values so I can't give you frequency of de-essing but:
Just after the f/b transformer are two poles of the appropriate switch and, in De-S mode, one pole unshorts two fixed caps (120p & 33p), one adjustable cap (5p?) and a resistor (?K).  The other pole connects a resistor (5K)

These components, now in the feedback path, form a fixed Hi-frequency shelf boost;  The first switch pole (with the 3 caps and one resistor) is the series arm of the filter and the other switch pole (with the 5K resistor) is in the shunt leg of the filter along with the already present 56K resistor.  Since these components are in the feedback path, the boost increases the sensitivity of the hi end, thus forcing the unit to compress them more.
In other words, the compression action is the reverse of what the filtyer does. 


Now that I read this back it may not make sense (I hope it does) but I'm tired so it might be the best I can do right now  :eek:
Good luck  :)
 
Many thanks Doug and Jean,

I've got a more specific enquiry, we're trying to fathom out what is happening with the anode supply to the 2xSK7, where there are 2x56K paralleled by 3x100K resistors joined in the middle by a 2uf cap?

(apologies if there is any senselessness here, it's all my fault, as a non technical lump sitting between the questions my dad is coming up with and the internet!)

He's also spotted what he thinks is a family resemblance between our circuit and the UA175B (as posted here: http://www.tangible-technology.com/dynamics/comp_lim_ec_dh_pw2.html ), is this just because they're both feedback vari-mu topologies, or do we stand to learn more by looking at the 175 in a bit more detail?

Cheers,
Child
 
Hi there,
the network you're describing for the 6K7 anode supply is the DC "balance" for those valves.  And the network in the following 6BR7 grid circuits (2 X 56K + pot) is to balance the AC signal coming from the 6K7's .  You'll want to balance everything as best you can when in "Bal" mode to avoid "thumping" effects etc.  I daresay it'll be a juggling act with both pots until you get the optimum balance.

IMO,  any similarity between your unit and a Universal Audio 176 is as you say, purely because they're both vari-mu/bias topologies.
I bet Doug would know if there's a closer circuit to yours for something to study, he's well versed on vintage compressors of this type.
However, they all work in a similar manner, give or take, so your Dad and yourself might find any of a similar vintage useful or interesting to study regardless.
I'd say that your circuit is quite comprehensive compared to some which are quite basic units designed for simple or brute force duty.
Also, I don't know what the original application was for your unit but, as you may know, certain vari-bias compressors were designed with restricted bandwidth transformers for speech/communication use.  I'm not saying yours was, but you may want to measure the response sooner than later before you invest a ton of time or money.  You never know though, you might have a great unit in the making.


Cheers.


P.S.  I recognise some of the British brands of parts underneath from my youth... nostalgia trip 8)

 
Jean, thank you so much.

It's impressive to note that one forum post has earned you at least as much respect from the old man as I've managed in 35 years! ;-)

I think the main reason he took on the project was because he also got that wave of nostalgia. I'd previously been almost completely unsuccessful in enticing him to get involved in helping me with any of the cool projects that spring up in this place but he leaped on this one.

I wonder if you might be able to offer an opinion on the blue pinout on Jones plug 2? We're wondering if it might be intended to lead to a stabilised valve in the power supply, such as a VR105/OB2, or if it might just be a monitor point?

EDIT: Scratch that last question we've worked it out (it's a monitor point). Next up butchering the range of obscure devices I've retrieved from skips over the years to lace up a power supply!

Thanks again,
Child
 
JP2/blue at a 2 second glance is the metering point. 

There are 3-4 other RCA Photophone limiters that are the next closest thing, none of that data is online that I know of. 
 
Hi Rothchild,
hopefully you'll get the unit going with a power supply cobbled from what you've found dumpster diving and things'll fall into place without hassle.

Rothchild said:
...one forum post has earned you at least as much respect from the old man as I've managed in 35 years! ;-)

That's nice of you to say.  I was really just the first one to the plate though and, if you decide to stick around here, you'll see that there are quite a few folks around who are knowledable about all sorts of cool kit.

Anyway,  in turn, you can let your Dad know that his reversed schematic is a nice one.  :)

Cheers, good luck and have fun.
 
My first thought was that the RCA meatball was added later. I grew up near RCA broadcast and that chassis does not look "RCA" to me. But Doug suggests it is british made, which makes sense to me. The circuit is not un-RCA. RCA operated worldwide.

I'd love to have an RCA meatball again but I suppose they are now rather more precious than I would spend. (Long ago there were meatball tie-clips, cuff-links, etc... are they collectable today?)

> he thinks is a family resemblance between our circuit and the UA175B

The vary-gain input and the output stage are similar. Mostly/all vari-Mu schemes are. Differences are interstage coupling (R-C or transformer) and another gain between vari-Mu and the final. (Or as an extreme: Fairchild beefed-up the vari-Mu to also be the final audio stage.)

The simple jobs like uA175 drove the sidechain rectifier right off the audio final. It works, often very well.

This RCA takes sidechain through a 12AT7 gain-stage and AT7 cathode follower. This allows sidechain level to be different from audio output level. It also seems to be a place to eSS-boost the sidechain so that SSpitting speech syllables are ducked more than average speech. Partly for actors who hisss their essses, but also for mikes and loudspeakers (and optical film sound) with ratty highs which over-emphasized speech essses.

What is really over-the-top is the balance metering. If I am reading this right (?) it has pentode pentode triode triode gain, or something like 400,000. In BAL the gain-control line is switched to "JP2", an unspecified source. Other boxes injected a small 60Hz hum here. You balance trim the 6K7/6SK7 cathodes, plates, screens for minimum output. This gives first-order optimization for thump rejection. (A fast-acting audio limiter not perfectly balanced will "thump" as DC levels change quickly to control excess level.)

No, I don't know why there's balance pots in both plate and grid networks. The plate pot doesn't trim the DC case because of the 2uFd cap.

Interesting that I don't see a compression meter. How do you know you are hitting it too hard? (Perhaps the source meter reading is specified, and this box only takes-up the uncertainty of VU meters and/or momentary accidents.)

The interstage 0.025uFd against (56K||150K)+56K seems to be 3db down at 63 CPS (Hz). This may be more a newsreel channel than, say, NBC Orchestra. This too may be thump control.

0.01u+0.02u against 2.4Meg gives a 72mS release time, quite fast for music, more suited for avoiding speech overload. Attack time will be much faster, and suggests protection for a recording medium which clipped violently.

Doug mentioned PhotoPhone. This became a tradename not a technology, but RCA was known for ribbon shutters and some of them were embarrassed by thump or even momentary clipping.

6BR7 is an odd choice for "output" bottle. It is a small pentode for low-level audio. Here it is triode-wired, which apparently makes it similar to 6C5.... except 6BR7 is rated 0.75W plate and 6C5 is 2.5W plate. BTW I doubt the shared cathode resistor here is 68K; 680 seems more likely. Perhaps red/orange/brown stripes have faded over the decades?

6BR7 may have been one of "THE" tubes always in-stock around UK commercial audio maintenance facilities. (We see a lot of 6J7/1620 in US broadcast because it was the go-to tube from late 1930s into the 1950s.)

My data for 6BR7 suggest B+ may have to be <200V or the 6BR7s are past their plate dissipation. (Or the cathode bias really is 6.8K; but then current is awful tiny.) Output may overload by +20dbm. With no NFB around the output stage, THD may be high before +10dbm. This is well short of US broadcast studio limiters and their +36dbm line-slappers. However as a remote unit with everything in one truck it makes perfect sense.
 
PRR said:
But Doug suggests it is british made, which makes sense to me.
Plessey and what I recognise as  Hunts branded capacitors are utilised and both are common for British made units. 

PRR said:
No, I don't know why there's balance pots in both plate and grid networks. The plate pot doesn't trim the DC case because of the 2uFd cap.


Missed the cap.  My bad.
Equalising of DC bias on the .025uF coupling caps?  There's a similar (but more elaborate) schema of a British Broadcasting  limiter  from around 1965. It's very extensive in its adjustments for triming out the usual ills of blocking, thumping, distortion null. 
It also suggests to me that the 6BR7 cathode bias here is not 680R but is more of a current source.  Don't quote me on it though and it's easy to measure.

The TC of the inter-valve RC coupling in this case could also be an issue of safety rather than frequency.


Having no on-board meter (and what appears to be no remote connection for one) doesn't necessarily suprise as the unit could merely have been in-line as a safety device.  If it was part of a camera setup, a cameraman in a remote situation wouldn't be watching audio meters.  It wouldn't be his job to and might, depending on the particular organisation or country you were in, even cause a "walk-off"  if he did.


Edit for removing some irrelevant waffle.
 
Jean, Doug and PRR

Many thanks for the input, this discussion is invaluable to us, please keep it coming!

The mid 60's dating seems right to me, I've got this rather poor shot of one of the big caps that appears to say 'May '63' on it.

http://www.miwi-music.com/MusicAndMedia/rothchild/rca/rcadateoncap.JPG

In a little more detail, the back story I got from the guy I bought it from was that it's from: "an RCA PM80 16MM optical sound recording camera, which was all blue, in three units, one of which was a compressor." which seems to concur with what you're all saying.

Thanks again,
Child

 
Rothchild,
I meant to say in my last post that, you'll get an idea of the value of the intended H.T. voltage from the working voltage value written on the smoothing caps. 
Also that, once you have the unit running and assuming it's worth it,  things like adding a meter for indication of compression are fairly easy modifications.
A typical method for this is a 200uA (or a standard VU with the diode bridge removed) meter that monitors the current change in the cathodes of the 6SK7's. 
Currently (Ha! an unintended pun) there's a 2K2 resistor in the cathodes but a resistive divider instead with an appropriate value to monitor a portion of the total current that taps off to the meter would be easy.  Maybe a pot to adjust for "zero" too.
I can't remember the exact setup used but the Universal 176 you mentioned is an example of this arrangement.

If the unit's output stage does crap-out at fairly low levels (as PRR postulated), it doesn't necessarily mean it'll be a no go.
Depends on what you're driving with it.  And I doubt it'll be a broadcast transmitter you're building so ?

Also, I remember being at a trade show and talking with Miss EvaAnna Manley.  The topic of some other manufacturers limiter unit came up and she mentioned one of the the main specifications being touted at the other guy's booth was "Big Dynamic Range!".
We laughed at the irony of the spec, given that the unit was supposed to limit the same thing.

Anyway, again good luck.




 
Jean Clochet said:
If the unit's output stage does crap-out at fairly low levels (as PRR postulated), it doesn't necessarily mean it'll be a no go.
Depends on what you're driving with it.  And I doubt it'll be a broadcast transmitter you're building so ?

I was hoping I'd be able to loop it in as an insert on my DAW (via an RME Multiface) do you think I'm likely to need some sort of amplification between the D/A out and the input of this box then, (and does that mean that I might need to pad it back down again before returning it to the A/D also)?

Cheers,
Child
 
Rothchild said:
do you think I'm likely to need some sort of amplification between the D/A out and the input of this box then, (and does that mean that I might need to pad it back down again before returning it to the A/D also)?

Given the use of this, it's possible that the through gain of the unit is at or close to unity.  There are already pads on the input as part of the ratio setting.  I don't know the exact ratio of the output transformer?  I could guess at the total gain if I had a bit more time to look at the doc, but I'm late. 

Ideally, you'll just need to determine the correct level of send from your DAW so that the ratios etc are working correctly and the output isn't distorting - hopefully the level you then get from the output of the RCA and back into your DAW will be about the same as the send but compressed.

I don't know where you set your digital reference and you may have to lower the DAW send if the RCA output stage is less than beefy, but that shouldn't be an issue. 

Have to dash...


Edit: Back again briefly.  I remembered where I'd seen that arangement for balancing before, it was on a Westrex compressor.  The schematic and a description is in the Audio Cyclopedia - the Tremaine version.  I don't have my copy anywhere nearby but maybe someone does and can look it up.  It might go into some detail regarding the thinking behind the two sets of adjustments.

 
Yeah, we went around on this piece at GS first, and I initially thought it was a home made piece someone had put an RCA badge onto.  It bears no physical resemblance to anything made by RCA in the US, and there aren't many foreign variations out there.  There are RCA UK Photophone theater amps that are very similar to the US models, and there's an Australian RCA remote mixer that pops up on ebay every few years; no idea what that thing is either. 

These units were all safety controls for cameras.  Anything RCA having to do with film on any level came from the Photophone division, which made many parallel yet different products from the broadcast division. 

The US counterparts are usually about 30 dB gain, that's with compression active.  Usually ratio is 1.5-4:1. 

 
PRR said:
My data for 6BR7 suggest B+ may have to be <200V or the 6BR7s are past their plate dissipation. (Or the cathode bias really is 6.8K; but then current is awful tiny.)

Well done PRR, I've just been jumped another notch in the pecking order. ;-) That 6.8k is actually 680ohm.

My dad says he's spotted a couple of other transcription errors and has already got a v1.1 of the diagram that he's working from. Is it worth submitting this to the database here once we're done and know it works?

Work on the power supply has started in earnest (using a transformer from an old oscilloscope I rescued from a dumpster a few years ago) I'll try and get some pics up once there's something tangible to show for it.

Cheers,
Child
 
Power supply components have been identified and sourced and work's started in earnest:

DSCI0086.JPG


Can anyone suggest where I might be able to source a pair (M/F) of these Jones plugs? (can anyone confirm or deny that they are Plessey 'Multicon' connectors?)

rcaio1.JPG


Cheers,
Child
 
Hi all, sorry for the late bump of this thread but work has continued on this box and it looks like we're good to starting testing and calibration this weekend.

I just wanted to share with you the finished schematic:

Circuit+diag.jpg


And see if anyone had any useful tips to help get this unit set up? I'm going to take my laptop and UA25 soundcard with a bunch of test tones, so we can set up determined levels, calibrate the input threshold and so on. Obviously I'll also have a fat pack of drums and some vox tracks to try some real world examples also.

Cheers,
Child
 
Back
Top