EMI RS61 Schemie

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

Winston OBoogie

Well-known member
Joined
Jun 3, 2004
Messages
2,520
Location
UK.
Here's another found schematic from my old sent email folder, I think this may have been posted but couldn't see a copy so...
Drawn hastily after getting under the hood of one a few years back, as usual, a few things are missing.
From notes, the anode choke on V3 is 72H and I see there's a ground connection missing.  You can figure it out.
A nice 40dB amplifier, gotta love those EF37's. 
 

Attachments

  • RS61 Amp.pdf
    34.7 KB · Views: 145
ruffrecords said:
Interesting negative feedback scheme.

Cheers

Ian

It's not seen.too often.  There's a 9 pin miniature BBC designed amp that returned fb via the earthy end of the transformer secondary but I'm struggling to think of any other.
 
Lots of that input transformer primary return path in textbooks, not too much I've ever seen in the field.  The dual path reminds me of the many super-tailored things Gates did. 
 
EmRR said:
Lots of that input transformer primary return path in textbooks, not too much I've ever seen in the field.  The dual path reminds me of the many super-tailored things Gates did.

YES!  You do see examples in the usual tomes for sure.  If anyone, I was thinking you would know of a U.S amplifier like this.

I think the dual path might almost be inevitable with 3 inverting gain stages enclosed in a main loop.  I also see both a lead (across the 1st stage anode) and a lag  (across the inner loop R) cap to keep things under control at the high end. 


 
Winston O'Boogie said:
YES!  You do see examples in the usual tomes for sure.  If anyone, I was thinking you would know of a U.S amplifier like this.

I think the dual path might almost be inevitable with 3 inverting gain stages enclosed in a main loop.  I also see both a lead (across the 1st stage anode) and a lag  (across the inner loop R) cap to keep things under control at the high end.

Indeed. It amazes me they were able to work all this out with little more than a brain and slide rule.

Cheers

Ian
 
ruffrecords said:
Indeed. It amazes me they were able to work all this out with little more than a brain and slide rule.

Yes.  Although there were still quite a few bright boffins at EMI Hayes during this period.

A rack of these amplifiers was used in the first stereo mixer which was the REDD.1 system produced in 1955 or so.
This is not that long after the founder and head of REDD, Len Page, had been working under Alan D. Blumlein.
Compared to Blumlein's efforts in radar up until his death, and EMI's continued work in that field as well as the EMI-Marconi TV system and various government computer projects, I expect a little ol' tube amp was something a lot of them could knock out in an afternoon 😃
 
Hey Ian, PRR, or anyone,
If you see this, see what you think of the 1st valve biasing.
Using the curves in the attached datasheet, a quick eyeball seems to put it at around 2.5mA Iq and about 3.75v on the cathode.  I assumed about 200V supply to the 27K anode load based on a 260V HT.

Seems like they were concerned about input overload given it's about twice the usual cathode bias for an input valve.
I also noted the less than usual 1:5 input transformer ratio.
Lastly, I got the amplifier gain wrong @ 40dB, looks to be between 48 & 54dB depending on adjustment.
Cheers
 

Attachments

  • EF37.pdf
    371.6 KB · Views: 18
Normally you would expect the negative feedback to mitigate this. However, the last two stage have local NFB giving a gain of about 34dB. The mu of the first stage as a triode is 28 so you are not likely to get much more than a gain of 20 times or 26dB out of this stage. Total open loop gain is therefore about 34 +26 = 60dB.

At the lowest overall gain setting determine by the global NFB to the first stage the expected closed loop gain is about 820K/1K5 = 546 = close to 55dB. So there is no more than 5dB of NFB due to the global NFB loop. So the NFB is not doing much in the way of mitigation so maybe they do need a higher bias.

Cheers

Ian
 
ruffrecords said:
At the lowest overall gain setting determine by the global NFB to the first stage the expected closed loop gain is about 820K/1K5 = 546 = close to 55dB. So there is no more than 5dB of NFB due to the global NFB loop. So the NFB is not doing much in the way of mitigation so maybe they do need a higher bias.

Ah, I forgot to check the gain of the other two stages with their closed loop but without the global so, yes that makes sense why they biased high.  Looks like the global loop is only for gain trim then, no great reduction of THD, although knowing how they generally designed these things, probably some cancellation between 1st stage and other 2 together.
Anyway, thanks for checking it out Ian, much appreciated 👍

 
Found this attached to an old email.  I had a few missing part values when I drew my schematic and eventually ended up trading drawings with a guy who'd also had a unit to inspect.
 

Attachments

  • RS61.jpg
    RS61.jpg
    151.3 KB · Views: 132

Latest posts

Back
Top