[ ultra newbie ] adding switchable transformers BBC AM7/4B line amps

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

Dinosaw

Member
Joined
Feb 25, 2019
Messages
10
As the title suggests, I'm an ultra newbie, the most complicated electronics project I've done is wire a stratocaster with a diagram or replace the speakers in a guitar amp. Please let me know if there is a more appropriate newbie section where I should post this.

Sooo I bought these germanium pres that the BBC made in the 60s - balanced in and out. I'm unable to read the schematic here:

http://www.bbceng.info/ti/eqpt/AM7_4.pdf

but I can see that the input impedance is 600 ohms and the output is 55 ohms.  The pres sound great but they aren't very high bandwidth and their noise floor is on the higher side.  I thought maybe I could add very high quality input and output transformers to get more bandwidth, then make them switchable so I could switch between them and the stock ones like an eq lift.

I would have this done by a professional but as I go I want to understand something about the process so I have a few questions.  I was on Sowter's webpage looking at input transformers and noticed the 1364 EMI TDS5383 input transformer - it doesn't specify whether it's mic or line - could it be both?  It says it's a replacement for an EMI germanium console and can be tapped at 600 ohms, which is the input impedance of the BBC pres I bought - does it mean that I can use it? 

http://www.sowter.co.uk/cgi-bin/sh000001.pl?WD=emi&PN=VINTAGE-RECORDING-AMPLIFIER-TRANSFORMERS-A-to-K%2ehtml#a530

There are input and output transformer sets that would seem to match the impedance of the pre, such as those made for variations of the API 312 - is it possible to just pick transformer combinations out of the blue like that if they match my pre's impedance?

Last question, I noticed that cinemag does output transformers that are humbucking - could this help reduce the noise floor of the pre itself or does it only cancel our hum that comes from external interference?

Thanks so much for your help.
 
Well before you change anything you should MEASURE! Measure the frequency response of the first transformer. Then measure the frequency response of the whole unit. You could also measure separately the output transformer. If you dont do this, you have no idea how good the transformers are. BBC gear was very high quality, but bandwidth was limited in some cases to 15 kHz. There are a number of ways you can do this, not necessarily just the transformers.
 
Hi!  And thanks so much for replying, I can't even get anyone to talk to me about this and obviously I have no idea what I'm doing, so I really appreciate you taking the time to comment.

Okay so I guess I'm thinking about circuits incorrectly.  I had this idea like transformers were at the beginning and end of the circuit and acted as gated keepers of bandwidth, so changing them would change the bandwidth as well as the character of the pre.

So okay let me try to understand - there's two things that could happen now. I could measure the BBC transformers and find that they actually have amazing bandwidth, rivalling the sowters I'm talking about.  From there I would just need to figure out how the sound is being filtered in the middle of the circuit and make some of those (much less expensive) components switchable to allow greater bass and treble response.

Or I could find that the BBC transformers are indeed limiting the signal - -my hypothesis is that they are limiting both the treble and bass response...but we'll see.  Then I could go ahead with a recap of the unit as well as switchable transformers?

I would really like to know regardless, where I could learn about how to match potential transformers to a circuit and how one does this - I don't even know how to word the question to search for an answer online - do you know of materials I could use to learn this?
 
Specs says response at 15KHz to -0.3dB - which actually is pretty good, and nothing I suspect could be significantly improved by swapping transformers. BBC were never cheap with their iron.

Not that you can get anything that drops in for the output transformer, it's rather specially wound for quasi-complementary transistor output, DC tolerant, and emitter feedback windings. Except maybe (maybe) a Neumann BV157/BV167 from a PV76 - but those don't come cheap.

Your noise floor is from using Germanium - not much you can do here either

Check that you get decent/predicted/approximate  response for noise and frequency range, so you can exclude a damaged component.

Jakob E.
 
Thank you so much for responding - I know that you and many of the contributors here are legendary gear builders and you have better things to do with your time! It seems that the only option is to get this pre in spec, let it do what it does well, then for applications that require a different frequency response get a different pre...which will unfortunately be even more costly than more transformers haha

So I won't pursue this project, but just so I understand the concepts involved - with the example I gave in the first post of the sowter 1364 EMI TDS5383 input transformer - all that's listed next to the part is impedance.  If it can be tapped at 600 ohms, which we know the pre needs and we know that it's optimized for a germanium circuit, what else do I need to know in order to determine if it's a match for the circuit?  All I'm trying to do at this point is walk away with some greater understanding of this topic that I can go research to death in my own time.
 
I think almost all the bbc amps were limited in bandwidth to 15 khz back in the day , mostly it was AM transmissions and even FM doesnt go much beyond 15 khz anyway. Maybe all those  470pf caps around the input bleed off some treble to ground , to keep the feedback loop stable . At the expense of stabillity you may be able to gain some small extra bandwidth . probably best to leave things as they are ,or at least put everything back right and just accept the slight loss in the highs , keep them for sources that dont require much top end .I think your transformers arent the limiting factor in this unit , the circuit is intentionally limited in bandwidth , c1 c2 c 13 and c14 seem to form a low pass filter.
 
This is not a "pre". It is a line receiving amplifier. It takes a fairly high-level signal from a long lossy line (think London to Birmingham) and restores the level, also giving enough output to drive *another* amp (line sending) to put onto another long line (think Liverpool).
http://downloads.bbc.co.uk/rd/pubs/archive/pdffiles/monographs/bbc_monograph_26.pdf

The noise will be horrid unless coarse gain control is on the max-gain setting. At this point the gain will be too much for most uses. (This setting is for extremely long lines followed by heavy passive EQ to correct treble loss, a thing commonly done on very long lines for telephone or AM-grade radio links.)

The input transformer, they say, is tuned for low-pass at 50kc/s (50KHz). There's more treble response than many ADCs know what to do with.

The output small-signal impedance is rated 55 Ohms but it certainly is not meant to *drive* 55 Ohm loads. They want you to measure it with 15K load.

Under that condition, the output for "serious distortion" should be "+17dB". Reference is not stated but perhaps 5V-7V.

Noise Factor is said to be "7dB to 9dB", which is not horrible for high-level mikes but a $10 chip will do 6dB better. That's on the top-gain stud which is likely to overload easily in modern studio practice.

Rack 'em for looks but don't bother to patch them up. They are not what you want. Any "improvement" would amount to wholesale re-design and part replacement.
 
I wouldn't want to put anyone off these units racked as pres.  Their noise floor is noticeable to be sure but they have a lot of color and a lot of unique character for anyone who really does appreciate a 'vintage' sound.  The reason I wanted to expand on them was more because I like them rather than them being inadequate.

In any case, I really appreciate all of you guys helping me analyze the circuit, I'm learning a lot not only about 'the way things work' but also the history of analog gear!
 

Latest posts

Back
Top