Lynn Olson Raven Question

[zelter]

Hello,

does anybody know the voltage amplification factor of this linestage? Has someone build a Raven? Recommendations and comments are welcome.

Zelter

[zigzagflux]

http://www.nutshellhifi.com/Raven-MarkII.gif

A fine design. Very happy with the sound. Highly recommend Lundahl amorphous LL1689 for the output transformer, and give it about 25 hours before you judge. I use a current output D/A to an I/V resistor stage, which then feeds the grids of the input tube.

When measurements are taken differentially, the gain will be the mu of the triode divided by the turns ratio of the output transformer. Very simple calculation.

[Looneytunes]

Hi zigzag,

I've been thinking about building the Raven for a long time. Can I ask which tube you are using? The ECC99? And what capacitor are you using at the CT of the output transformer?

Thanks,
Looney

[zigzagflux]

I use the 6n6p at 110V plate. I think around 21mA per triode, but would recommend a little more (somewhat based on whatever headroom at the grid you can handle). This tube likes its current. The capacitor is a motor run bypassed with a film/foil polypropylene. Would recommend trying out the inexpensive FT-1 teflon caps as a bypass instead.

This build dispenses with the input transformer, and uses a balanced stepped attenuator as a volume control. Works out pretty well; I get 1.7V at 0dB with the RAKK DAC II and 75 ohm I/V resistors. If I ever need less output, I can reduce the output turns, and get a bonus of lower output impedance. AC heating on the tube; originally used DC heating but this actually caused more problems than it solved. I would suggest using dedicated transformers for both the rectifier and the triode. The power transformer is ideally shielded. If I could do it all over again, I would ask Jack at Electraprint to build the power tranny. I would also take the time to orient the output transformers for minimum stray field pickup; I get just a tad of pickup that I neglected to consider ahead of time. But nothing you can hear with your ear; just viewable on the scope.

[Looneytunes]

Thanks for the great info, zigzag!

Sadly, I won't be able to start my I/V project for a few months. I have a couple projects in the fire at the moment (bad pun) that I need to finish. But, you've definitely given me a starting point.

Can I ask one more question? When designing your circuit, did you comapre Lynn's Raven approach (CCS at the CT) to Dave Davenport's approach (two CCSs on both sides of the transformer)? I've always been curious about the trade-offs between these two approaches.

Thanks again,
Looney

[revintage]

I am atthe moment building a variation of Raven mkI with LL1667(PP-choke intead of two separate), LL1676(2+2:1) and a Mundorf silver at the output. I use the the russian 6H30 that I love.

I recommend that you take it one step further and use a balanced output TVC at the input. I use the 23-step Silk Super Premalloy.

zigzagflux: Do you have the story behind going from MkI to MkII?

[Looneytunes]

Hi Lars,

Thanks for all of your suggestions. This is the first time I've seen the schematic you posted! I can't believe that I missed it. Where did you find it? Is there a thread out there where Lynn discusses his design? Also, do you have a higher resolution schematic, or a link to one? I can't read some of the smaller notes. (I did find one on the Sowter site, but the resolution is terrible.)

Also, do you have any comments regarding the Sowter transformer vs. the Lundahl?

And finally, do you know which schematic is the latest? I would guess the MkII, but I'm a little confused about the dates.

Sorry about all the questions, you can tell I'm excited to find another Raven.

Thanks,
Looney

[revintage]

Would never substitute a Lundahl for a Sowter. Maybe a little biased by the fact that Lundahl is also Swedish.......

Go to www.sowter.co.uk and search for the transformer and you will find the schematic. Not sure though that solution is better.

Off course MkI must be the older one.

[ErikdeBest]

Hi Looney

Here is the link for the sowter based Raven.

http://www.sowter.co.uk/specs/9111.htm

I came across it a long time ago, by mere coincidence. I do not know why it is not published or described at Lynn Olson's website. Maybe he liked the Lundahl based Raven better?

[ErikdeBest]

Can I ask one more question? When designing your circuit, did you comapre Lynn's Raven approach (CCS at the CT) to Dave Davenport's approach (two CCSs on both sides of the transformer)? I've always been curious about the trade-offs between these two approaches.

Hi Looney, hi Lars

About the CCS at the CT. I know that from the last schematic at this page , also by Lynn Olson. The Lundahl based Raven has indeed got a CCS to feed the CT, but it also feeds a voltage reference tube. I think that due to this voltage reference tube one can't speak of a true CCS feeding the CT. The voltage reference tube is shunting some of the current to the ground, and it does vary the amount of current in order to keep the voltage stable. That implies that the current fed to the CT is not constant and neither provided by a high resistant source. But that is just my impression, and probably I am wrong. I just can't get the idea of how it is working!

Erik