Redd. 47 preamp + Redd EQ

[ruffrecords]

So is there no type of attenuator that would work before the output transformer primary?

Not really no. That transformer and output stage are designed to be connected directly together. It is not a power match but a voltage feed where the load impedance is much higher than the source impedance. This is generally the case inside an amplifier or mixer. Today it is also the case outside the mixer. In the old days a power match was used which lost 6dB of signal in every matched connection.

If so am I going to need another set of transformers to interface the eq properly with the amplifier?

You only need another set of transformers if you want to run the EQ balanced. There is no reason to do this. You could just drive the EQ direct from the ECC88. You could feed the output of the EQ into a step down transformer to make the output balanced and low level then feed it into the the input transformer of another REDD 47 just like in the original desk.

However, if you want the final level to be line level without using another complete REDD 47 then you need a gain make up stage for the EQ followed by and output transformer.

Cheers

Ian

 

[emi2345]

Not really no. That transformer and output stage are designed to be connected directly together. It is not a power match but a voltage feed where the load impedance is much higher than the source impedance. This is generally the case inside an amplifier or mixer. Today it is also the case outside the mixer. In the old days a power match was used which lost 6dB of signal in every matched connection.You only need another set of transformers if you want to run the EQ balanced. There is no reason to do this. You could just drive the EQ direct from the ECC88. You could feed the output of the EQ into a step down transformer to make the output balanced and low level then feed it into the the input transformer of another REDD 47 just like in the original desk.

However, if you want the final level to be line level without using another complete REDD 47 then you need a gain make up stage for the EQ followed by and output transformer.

Cheers

Ian

Ok I think I will go with the option of inserting the EQ between the E88CC and output transformer, and use a gain make up stage, provided you think the input impedance of the EQ at all settings is high enough for the E88CC to drive it. Since I want to be able to switch the EQ in and out of circuit, the output of the EQ would go to the same output transformer (Sowter 9980 30k:600), what would be the best type of gain make up stage? Simply another parallel E88CC? Or perhaps a 12A_7 using one half as a cathode follower before the EQ and the other half as common cathode gain stage after it?

 

[wmarden]

This is a great idea for a project. I like the idea of having inserts pre eq so you could comp pre or post.

 

[ruffrecords]

Ok I think I will go with the option of inserting the EQ between the E88CC and output transformer, and use a gain make up stage, provided you think the input impedance of the EQ at all settings is high enough for the E88CC to drive it. Since I want to be able to switch the EQ in and out of circuit, the output of the EQ would go to the same output transformer (Sowter 9980 30k:600), what would be the best type of gain make up stage? Simply another parallel E88CC? Or perhaps a 12A_7 using one half as a cathode follower before the EQ and the other half as common cathode gain stage after it?

The REDD EQ canb e driven from a 10K fader so the ECC88 will have no problem with it. THe REDD EQ needs 10 to 12dB of gain make up so a repeat of the ECC88 output stage should be fine. Don't forget to include the 330K input resistor.

Cheers

Ian

 

[emi2345]

Great thanks Ian. What effect does taking the E88CC out of the negative feedback loop have on its output impedance? It will be  a bit higher? Does this parallel E88CC plus 7:1 transformer make a good general purpose low impedance driver? I’d like to fit a balanced low-z tube buffered output to my ampeg b-15 build

Thanks,

Edmund

 

[ruffrecords]

Great thanks Ian. What effect does taking the E88CC out of the negative feedback loop have on its output impedance? It will be  a bit higher? Does this parallel E88CC plus 7:1 transformer make a good general purpose low impedance driver? I’d like to fit a balanced low-z tube buffered output to my ampeg b-15 build

Thanks,

Edmund

Yes the E88CC stage will have a higher output impedance as it is no longer within the overall feedback loop. However there is some local feedback.  The original has a 7:1 output in order to feed a 200 ohm load. For a 600 ohm load you only need a 4:1 transformer.

Does it make a good general purpose low impedance driver? Output stages need two things: low impedance and drive capability. Drive capability is set principally by the quiescent current which is one reason EMI used a paralleled pair of triodes. Output impedance is set by the topology and negative feedback. So this stage, on its own, certainly has the drive capability but its output impedance is not as low as it might be,

The nominal closed loop gain is set by the ratio of the 1M6 and 300K resistors and is about 14dB. The nominal output impedance is the parallel plate resistance of the two E88CC tubes which is around 1K. The open loop stage gain will be at least 20dB so there is at least 6dB of NFB which reduces the output impedance to about 500 ohms. Via a 4:1 transformer this becomes just of 30 ohms which is plenty low enough for a 600 ohm load.

So yes, it does make a good general purpose low impedance driver.

Cheers

Ian

 

[emi2345]

Great, I understand it much better now. So if I were to buy the Sowter 9980 transformer which is 7:1+1 and wire the secondary in series that would give me 3.5:1 which is close enough?

 

[ruffrecords]

Great, I understand it much better now. So if I were to buy the Sowter 9980 transformer which is 7:1+1 and wire the secondary in series that would give me 3.5:1 which is close enough?

Yes that will probably be fine.

Cheers

Ian

 

[emi2345]

Great this should be an interesting start before I build the actual preamp. I don’t really want to drill another hole in the chassis though, can I put the E88CC inside an unventilated 1U 19”x10” steel enclosure, or will I get heat problems?

I can't fit that chunky Sowter 9980 inside the chassis, would the 8540 do instead?  It is 24mm tall so would fit

http://www.sowter.co.uk/specs/8540.htm

 

[ruffrecords]

Great this should be an interesting start before I build the actual preamp. I don’t really want to drill another hole in the chassis though, can I put the E88CC inside an unventilated 1U 19”x10” steel enclosure, or will I get heat problems?

You will not get heat problems in an enclosure that large.

I can't fit that chunky Sowter 9980 inside the chassis, would the 8540 do instead?  It is 24mm tall so would fit

http://www.sowter.co.uk/specs/8540.htm

Probably not - it is an input transformer.

Cheers

Ian

 

[emi2345]

Oh I thought it doesn’t really matter which way round you wire a transformer. The Sowter 4383 says it can go either way round would that be better?

http://www.sowter.co.uk/specs/4383.php

 

[ruffrecords]

Oh I thought it doesn’t really matter which way round you wire a transformer. The Sowter 4383 says it can go either way round would that be better?

http://www.sowter.co.uk/specs/4383.php

It is not the way round you wire it, it is the fact that it is an input transformer so it is not designed to handle the power needed for an output stage - which is one reason it is smaller.

Cheers

Ian

 

[emi2345]

Ok I’ll try and find an appropriate 4:1 output transformer that will fit. Any suggestions would be much appreciated.

I have another question about the 9970 input transformer. It has a 35% secondary tap. Does this mean I could wire a switch to change the transformer ratio to 1:2.45, for use as a line level input?

 

[ruffrecords]

Ok I’ll try and find an appropriate 4:1 output transformer that will fit. Any suggestions would be much appreciated.

Carnhill do some 9K6:600 transformers. Check out audiomaintenance.com

I have another question about the 9970 input transformer. It has a 35% secondary tap. Does this mean I could wire a switch to change the transformer ratio to 1:2.45, for use as a line level input?

Probably not. The problem is the input transformer is not really designed to handle line levels so even though the ratio is much reduced is may well begin to saturate.


Cheers

Ian

 

[emi2345]

The CA-18-VTB2291 looks like the one, but will it induce hum if I put it inside the steel housing of my amp near the input jack (picture attached)? It doesn't seem to have any mumetal shielding.

http://www.audiomaintenance.com/acatalog/CA-18-VTB2291_extended_info.html

With the REDD, I would like to be able to run line level signals for EQing as well as using it as a mic preamp, I was hoping the 20db pad and reduced transformer ratio would be enough, but perhaps it's optimistic to get it working well for such a range of signals.

 

[ruffrecords]

Both Holger and I have noticed that the big Carnhill output transformers can pick up small amounts of hum. We both tend to use Sowter now because they are screened but you are talking three times the price of the Carnhills. I have used Edcor XSM output transformers on occasion and they perform as well as the Carnhills but, despite not being screened, they do not seem to pick up any hum.

I forget the minimum gain from the REDD 47 but if you switch in a pad at the input equal to the gain then you have a unity gain line amp. A couple of 4K7 resistors and a 150 ohm will give you a 36dB attenuator and a 10K bridging input.

Cheers

Ian

 

[emi2345]

I built this unit and have been using it with great success for a couple of years. I went for a no expense spared approach with the sowter redd transformers. It sounds excellent and I’m very happy with it, many thanks Ian and everyone else for help with it. I’ve put on a couple of photos for you to see.

One tweak I’d like to make is to adjust the gain of the eq make-up amp circuit to unity, it’s a little low at the moment. I thought the best thing would be to add a trimmer to adjust the local feedback around the ECC88. Should I adjust the grid resistor (330k) to lower its value or the feedback resistor (1m6)? I’ve attached the relevant part of the schematic with my current voltage measurements, for both channels, yes one of the tubes is a little dodgy and needs replacing. My unventilated enclosure was overheating a lot so I’ve put a perforated aluminium lid on it instead now.

Many thanks,

Edmund

 

[ruffrecords]

Excellent build and a really smart front panel. Is that 3U high?

I like the ribbon cable wired directly to the Grayhill switches. Pity nobody does an IDC rotatry switch.

Cheers

Ian

 

[emi2345]

Thank you Ian, yes it’s a 3U schroff enclosure I had from farnell, and I designed the front panel using front panel designer, I sent it off to Schaeffer AG who milled and powder coated it, it was surprisingly easy actually. I leaned heavily on the design of the redd desks and the brilliance eq with the grey finish and the black knobs. Emis are my initials hence the cheeky alteration.

Any ideas about adjusting the local feedback around that eq gain make-up and output driver circuit?

 

[ruffrecords]

The gain is basically set by the ratio of R101 and R100. At the moment it is about 1200 divided by 330 which is about 11dB. If you raised R101 to 2Meg this would become about 15dB. I would not try to raise it any higher because you would be lucky if the stage had 30dB of open loop gain.

I would not advice lowering the 330K instead because the EQ really needs to look into as high an impedance as possible.

Cheers

Ian