R*E*D*D*ish mic preamp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

Val_r

Well-known member
Joined
Jun 5, 2004
Messages
306
Location
Naples, Italy.
Hi,

Just a few changes:
Added a line pad circuit,
constant current sinks biasing both tubes,
different working points,
changed E88CC with 6N1-P.

Please let me know what you think.

Thanks,

Val

:roll:

reddish.gif
 
just some thoughts:

input output transformer is the same? does the hammond transformer provide internal shielding?

missing b+ decoupling resistor between triode - pentode stage.

maybe you have to increase r11 like in the original to prevent the ef86 beeing loaded down too much.
 
Looks good!

In my pre (it is completely different, though) I am also using 6n1p and like its sound much more than ECC88/6DJ8.

A few things:
1) The Ri of the 6N1P is listed as 11K. In the real life it is probably somewhere around 20K, so if you are using 8:1 output trafo with tubes paralleled your output impedance is somewhere around 150 Ohm. Do you need it that low and lose 6db? In this configuration something like 4:1 might work better.

2) Try a single triode in the output stage (just short the grid, anode, and cathode or the second half) and listen. You might like the sound better.

3) I'd try to experiment with R13 value somewhere in 20K region. Actually, in my pre I am using a choke there, which to me sounds much better.

4) I am not a fan of changing amount of NFB as a volume control. In this case with every change you actually get a different sound.
Besides, I always like an option of no global NFB, for a different flavor.

I'd also experiment and listen how it sounds without R11 and R12 and also with connnecting the "cold" end of the output xformer to the cathode of the output stage.
 
Coupla tings:

Looks like you used my drawing as the basis for this. If so, bear in mind that the IP trannie secondary network of 20+pF & 33K is on there for use with the particular trannie I designed for use in a REDD.47 amp and also subsequently used in some Universal Audio pre-amps. It's now available off the shelf as the CMMI-7C and if you use this trannie then you're OK with that network, otherwise - A.O.T. (adjust on test)

I'd C.T. ground that shunt R for the Line Input.


The sound of a REDD is due in part to the regulated H.T. so use some type of reg there if you want 'that' sound. I've done it with neon's as in the original and also with a zener string on a MOS gate. Just built a unit with 'Gary-Pimm-like' C.C.S.'ed neon's and am digging it. Pick your favourite reg topology and, again, A.O.T.

Is your combination of R11 & R12 doing anything? The value of R11 is too low to be aiding the load on V1 and, shooting from the hip on the open loop gain of a 6N1P, I'd say no feedback is happening around V2.
I.M.O. an E88CC can benefit from a bit of F.B. to lower distortion as it's not the most linear valve on the planet. Can't say I've measured a 6N1P to see if it's the same in this regard...?

If you haven't already - build it, listen to it and... A.O.T.

Ciao

:thumb:
 
BTW: The A Designs MP-1/MP-2 preamp uses the exact same tube configuration: EF86 input stage and 6N1P output stage. I suspect that one isn't too far from the REDD47 design, either. But I wouldn't know for sure. I actually have an MP-2, but apart from trying different tubes I haven't messed much with the interior. Sounds very nice.
 
[quote author="Rossi"]BTW: The A Designs MP-1/MP-2 preamp uses the exact same tube configuration: EF86 input stage and 6N1P output stage. I suspect that one isn't too far from the REDD47 design, either...[/quote]

I did some circuit re-work and the pcb layout for the newer version of the A-Designs pre. Same valves but not the same circuit as a REDD.
 
Indeed. Not long ago, this place seemed very much on the decline. But with some of the old familiar faces turning up again, it's feeling more like it used to. The quality of the discussions has certainly taken an upward turn in recent weeks. Now we just need to get PRR to drop by more often and it'll really be a party.
 
Edit: Forgot to say
[quote author="Marik"]
Besides, I always like an option of no global NFB, for a different flavor.
[/quote]

If R11 & R12 are at the stock values of 330K & 1M6 then we have about 2dB of global N.F.B. so we're 'almost' there :wink:

And:
[quote author="Marik"]
...also with connnecting the "cold" end of the output xformer to the cathode of the output stage.[/quote]

Yep, returning the AC loop directly to the cathode is a nice way to go here. Also, if you're getting rid of the global F.B. loop as per Marik's preference then you could also move the output coupling cap from the "hot" to the "cold" end of the trannie. <mini-tantrum>I don't wanna hear from anyone who says it makes no bloody difference where the cap goes! - you've either never tried it, you're deaf or you're system was too mediocre to pick up the differences! Besides, it made enough of a difference for Western Electric to opt for this method in the 1930's </mini-tantrum>

Having built a few dozen variations based, or sometimes only loosely based, on this mic-amp, I can honestly say that I wasn't too fond of how any of the squalid-state C.C.S biasing arrangements I tried along the way sounded. One's own mileage may vary of course...

Post Scriptum: Don't cry Keef :cool:
 
Hi,

Thanks to everyone of you for your suggestions!

[quote author="CJ"]looks cool. how does it sound?
or no build yet?[/quote]

No, I didn't build it yet, I'm still in the phase of paper and calculator.

Is your combination of R11 & R12 doing anything? The value of R11 is too low to be aiding the load on V1 and, shooting from the hip on the open loop gain of a 6N1P, I'd say no feedback is happening around V2.
I.M.O. an E88CC can benefit from a bit of F.B. to lower distortion as it's not the most linear valve on the planet. Can't say I've measured a 6N1P to see if it's the same in this regard...?

If you haven't already - build it, listen to it and... A.O.T.

Yes, I will experiment along this path (aot'ting R11&R12).

[quote author="ioaudio"]does the hammond transformer provide internal shielding?
[/quote]
Yes

missing b+ decoupling resistor between triode - pentode stage

Both tubes have an Ebb of 300 volts, so no need for that.

I will post the regulated psu schem asap.

Respect,

Val

:oops:
 
[quote author="SSLtech"]Winston!

(sniff!)

It's just so good to have you back! :green:

:thumb:

Keef[/quote]

Amen!
 
[quote author="Val_r"]
Yes, I will experiment along this path (aot'ting R11&R12).
[/quote]

I've built some where the max gain was higher than 46dB. You can't get much more than 46 out of this pre without changing these values. I think something like 2M2 or thereabouts for R12 allowed another 6dB with some 2dB 'in hand' for F.B.
Abbey Road Studios have a unit like this that I built for them a while back. Sonically I generally prefer leaving well alone with this amp though and would rather add gain downstream with something else.

I suppose it depends on what your objective is with this. Starting a design from scratch, I doubt I would choose an E88CC or any of it's derivatives - I'm not a big fan of it's harmonic spread when naked which is why I suggested keeping the local loop from the REDD.47 on your design. I do like the JJ ECC99 though and would probably mess around with that in a non F.B. circuit.

[quote author="Val_r"]
I will post the regulated psu asap.
[/quote]


As with the audio path, to F.B. or not to F.B. is a hot topic of debate. I'm leaning towards simpler 'F.B.-less' topologies these days. Nothing wrong with the original reg circuit I.M.O. Buy enough valves and select for lowest hash noise though. Feed about 30mA to the pair of neon's with a C.C.S. and add a no bigger than .1uF cap across the output of them. it's a good place to start anyway and you can ALWAYS A.O.T. :wink:

Ciao

Post Scriptum: Did you see this page?

www.revolutionrecording.com/REDD47.html

I added a couple more pictures showing a bit of the internal stuff.
 
[quote author="Winston O'Boogie"]

I've built some where the max gain was higher than 46dB. You can't get much more than 46 out of this pre without changing these values. I think something like 2M2 or thereabouts for R12 allowed another 6dB with some 2dB 'in hand' for F.B.

Nothing wrong with the original reg circuit I.M.O. Buy enough valves and select for lowest hash noise though. Feed about 30mA to the pair of neon's with a C.C.S. and add a no bigger than .1uF cap across the output of them. it's a good place to start anyway and you can ALWAYS A.O.T. :wink:

Ciao[/quote]

There's enough for discussion here...
Changing OT from 8.15:1 to 4:1 will probably help a bit, what do you think?

[quote author="Winston O'Boogie"]Post Scriptum: Did you see this page?

www.revolutionrecording.com/REDD47.html

I added a couple more pictures showing a bit of the internal stuff.[/quote]

Yes I did. Happy to see the Real Thing !!

Here's my psu ...

Respect,

Val

:idea:

psu.gif
 
[quote author="Winston O'Boogie"]

I suppose it depends on what your objective is with this. Starting a design from scratch, I doubt I would choose an E88CC or any of it's derivatives - I'm not a big fan of it's harmonic spread when naked which is why I suggested keeping the local loop from the REDD.47 on your design. I do like the JJ ECC99 though and would probably mess around with that in a non F.B. circuit.[/quote]

I love the sound of the 6n1-P, and it's a military replica of the E88CC, I bought a bunch over the internet... as you can see...
6N1-P_bunch.jpg


and love the rugged construction of the Ulyanov plant (I think), see how beautiful is the grid wound!
(please note the space rocket, indicating the tube was for Air Force use, I believe 1972).
6N1-P_closeup.jpg


And more... it's a 300-volt rated tube, instead of the 130V of 6DJ8!

As you pointed out, the experimentation is sovereign here.

Respect,

Val
 
I see the current sinks as a whole lot of work just to avoid using two little resistors. Was the sonic change worth all that work?

The 6N1P is not a 6xxx derivative. I like to run them till they almost glow like a 6L6GC from back in the day. Hmmm.
And if you love 4.75Kwhat would you say about the 6H30.

analag
 
[quote author="analag"]I see the current sinks as a whole lot of work just to avoid using two little resistors. Was the sonic change worth all that work?[/quote]

I tried CCS in anode rail with a deplition MOSFETs and definitely heard a very nice sonic improvement. I did not use it though, as it was pretty noisey. Finally, I put there a chock and shut up.

[quote author="analag"] The 6N1P is not a 6xxx derivative. I like to run them till they almost glow like a 6L6GC from back in the day. Hmmm.
[/quote]

Yeah, they are pretty tough. 12ma with 200V on anode sounds about right.
 
[quote author="Winston O'Boogie"]
I did some circuit re-work and the pcb layout for the newer version of the A-Designs pre. Same valves but not the same circuit as a REDD.[/quote]

Wow, you're the man! :thumb: I like A Designs preamps. Unfortunately there's no German distributor anymore, so it's quite hard to get at the new ones.

Maybe you can help me with a few questions concerning my MP-2. For some reason my MP-2 has more gain than advertized, about 55 dB instead of 46 dB. Also, I discovered that the 250 V test point actually measures only 206V and the 6.3V heater voltage is only 6.0V. It's an European 230V version. My mains voltage was 225V at the time I measured. Do you think there's something wrong with my unit? It performs fine otherwise. Noise is a little bit higher, I think, than on an MP-1 I had on loan when they first came out (in Europe, late 2003, I think). But maybe that's just different tubes.
 
[quote author="Rossi"]
For some reason my MP-2 has more gain than advertized, about 55 dB instead of 46 dB. [/quote]

First, I'll clarify: I didn't design the A-Designs pre, just added some bells, whistles and do-dats to the MK 2 version along with laying out a new PCB.

Anyway, extra gain: You have a different IP transformer most likely. The later rev. switched this out so I'm assuming you have an early unit which also has this upgrade.

[quote author="Rossi"]
Also, I discovered that the 250 V test point actually measures only 206V and the 6.3V heater voltage is only 6.0V.[/quote]

Your 'lower than design centre' AC input would account for a slight drop in H.T. but the L.T. is regulated with an LT chip and thus wouldn't be affected - except for a dropping out of reg if the input were too low which I doubt it is. Nothing wrong with 6V on the heaters BTW, I usually put them there myself on pretty much everything wanting 6V3. Not sure where you're measuring the 206V H.T. voltage, I'd have to fire up my older computer to see the schemo's. However, I'm not sure that the 'test point' you mentioned shouldn't really read 200V - I'd bet your unit is just fine.
If voltage was an issue then you'd maybe be hearing a buzz noise due to the L.T. reg dropping out and you'd have potentially lower headroom or non-linearities due to some shifted operating points.

F.W.I.W. I swapped in a N.O.S. Mullard EF86 in a unit - microphoncs & noise dropped and sonics went up somewhat.

Hope that helps your peace of mind :wink:

Ciao.
 
Back
Top