Common Cathode as DAC output

[Arno Pf]

Well, preparations have started to convert the PCM63k based parasound 1500 DAC with an DAC 1:20 transformer, followed by a tube-stage.

For the latter I run into designs based on SRPP...but actually I would not need its amplifications due to the 1:20 transformer so a "simple" cathode follower E88CC would make more sense at a first glance.

Going through common cathode designs and tube-curves of the e88CC it seems that the low output impedance is excellent, however I am a bit worried if such a stage could "feed-through" the max 2,8V-peak signal that the DAC will be providing.

Any thoughts or experience on this that I should consider?

(By the way, I plan to reduce the total outputlevel bij decreasing the DAC-load, which will help its performance)

Cheers

arno

 

[funk1980]

I'm a bit confused by your post. Do you mean common cathode or cathode follower?

 

[Arno Pf]

I'm a bit confused by your post. Do you mean common cathode or cathode follower?

Seems I was also confused writing it....cathode follower it should be

 

[funk1980]

In that case, input headroom is not an issue. The gridcurves are based on the grid relative to the cathode. If grid signal rises, so does the cathode.

 

[Arno Pf]

Hm...in that case the advantage of an SRPP would only be the better linearity/lowTHD due to the "current source-like" function of the upper tube....right?

And....for SRPP cancelling 2nd harmonics is part of having low THD...(and maybe having a negative influence on "tube sound"....)?

 

[SY]

Actually, not even that. An SRPP is useful for swinging high voltages into difficult loads and the expense of some linearity. They are quite fashionable, but hardly ever used for where they actually would do some good. A plain jane grounded cathode stage buffered by a cathode follower will perform much better for a typical DAC stage which will swing a couple volts into the input of a preamp or power amp over cables that are less than 20 meters long. If the DAC plus transformer output is high enough, you only need the cathode follower, which will (if properly designed) have stunningly low distortion.

 

[Arno Pf]

Actually, not even that. An SRPP is useful for swinging high voltages into difficult loads and the expense of some linearity. They are quite fashionable, but hardly ever used for where they actually would do some good. A plain jane grounded cathode stage buffered by a cathode follower will perform much better for a typical DAC stage which will swing a couple volts into the input of a preamp or power amp over cables that are less than 20 meters long. If the DAC plus transformer output is high enough, you only need the cathode follower, which will (if properly designed) have stunningly low distortion.

That's what I aimed at. I planned to only have a 1x triode section behind the I/V transformer and have an as low as possible (non-transformer) output impedance. Power supply available of approx 200..220V and I also had in mind to parallel the two E88CC sections

 

[SY]

I also had in mind to parallel the two E88CC sections

Not much reason to do that- output impedance of an ECC88 single section CF will be something like 100R, which is quite low enough.

 

[Arno Pf]

Not much reason to do that- output impedance of an ECC88 single section CF will be something like 100R, which is quite low enough.

But on the other hand....If I want 1 E88CC per channel....why not put them in parallel and have extra low output impedance...

 

[Arno Pf]

So.... 2xe88cc parallel, Rk=10k...Vsupply=200V sounds like something that might work?

 

[funk1980]

But on the other hand....If I want 1 E88CC per channel....why not put them in parallel and have extra low output impedance...

You could, but you won't hear any improvements. So you could use just the one tube. Lower heater current and HT demand, simpler design, cheaper...

So.... 2xe88cc parallel, Rk=10k...Vsupply=200V sounds like something that might work?

When paralleled, 10k with 200V should be fine. Should be a 3 watt minimum though. Cathode bias resistor probably at around 180 ohms, depending on your preferred bias voltage.

 

[SY]

But on the other hand....If I want 1 E88CC per channel....why not put them in parallel and have extra low output impedance...

Why not 20 of them? Even lower!

Seriously, if that makes you happy, go ahead, but it won't change the actual performance in the system to any noticeable extent and will certainly cost you more in parts.

 

[Arno Pf]

Why not 20 of them? Even lower!

We'll if there would be 1 Tube with 10 double triodes in it...probably I would, but up till now a e88cc only has two..

It is because the choice is the E88CC and I do not favour to have 1 double triode being used for both channels....or using 2 tubes and leaving 1 triode-half per tube unused..

Hmm...reminds me of the 5842 by the way...strong little single triode tube...that's also in my 300B IT amp...

 

[SY]

I do not favour to have 1 double triode being used for both channels....

No electrical reason not to, just psychological.

 

[funk1980]

No electrical reason not to, just psychological.

+1
There are some excellent single tube buffer, headphone amps and other simple low gain amps out there! All stereo, all great performers.

But then again, the fun of DIY is to build exactly what and how you like it

 

[rhone]

With the 1:20 transformer do you need all the gain an ECC88 will give you? If not try a single 6H30P or 6N6P in a simple common-cathode circuit for an output impedance below 2kohms. It worked for me.

 

[Arno Pf]

With the 1:20 transformer do you need all the gain an ECC88 will give you? If not try a single 6H30P or 6N6P in a simple common-cathode circuit for an output impedance below 2kohms. It worked for me.

Required amplification by the 88 is only 1x so, don't need any amplification.

Just made the first set up and voltage checks. Now moving towards connecting the signal path

 

[kevinkr]

Here is my take on a tube and transformer based I/V converter..

Here: http://www.diyaudio.com/forums/tube...r-i-v-converter-differential-current-dac.html

Works quite well.

 

[Arno Pf]

Listening to it.......mellow....flow......excellent

Next steps....amplification vs DAC-load....tube rectifier....and reduce hum/rattle (must be the diode rectification HF torture that's flying around)

 

[Arno Pf]

Here is my take on a tube and transformer based I/V converter..

Here: http://www.diyaudio.com/forums/tube...r-i-v-converter-differential-current-dac.html

Works quite well.

Very nice...and using the 5842 as well....like the choke load...might move in that direction...thx!