Welcome! Need support, you got it. Or share youe ideas and experiences.
Return to K&K Audio / Lundahl Transformers
In Reply to: Re: Driver tube for 1660 posted by KevinC on October 31, 2003 at 05:11:29:
In the past couple of years, I have migrated to 3 stage amps to get away from the need for gain in the interstage transformer. This way I can use a low mu, low Rp tube for the driver and still have enough gain.I'm interested in some more details on this approach.
Are you looking for an input sensitivity of, say, 1VRMS, for full power while a direct source input from, say, a CD player? Should there be an active preamp (assuming an output voltage of 12VRMS) in the setup, we can go with a much wider range of input sensitivity. Is that right or did I miss something?
On the interstage transformer, would you go with a 1:1 transformer or simply no transformer in the coupling?
On the side, 5842 as a driver tube looks appealing with its high mutual conductance, high plate current, and lower rp. Yet just wonder if any in the octal or even DHT category would have the similar characteristics.
![]()
Follow Ups:
It's all about system matching and personal biases. I use a TVC normally for volume control, so I need a bit more gain than someone with an active line stage. Even the lovely sounding Raleigh Audio active line stage kit that I sell has relatively low gain (5dB) in its usual connection. However, it has provision for increasing the gain by reconnecting the input transformer for 6dB additional gain for a total of 11dB. My experience in retail audio taught me that most systems have too much gain, not too little, because manufacturers err on the side of more gain than less, because consumers can always reduce gain with the volume control. Unfortunately, that results in a noise and dynamics trade-off if carried too far, though.I would continue to use an interstage, but use it 2:1+1 or 1:1 for SE. Output tube drive is a critical part of the sound of power amplifiers. The use of wimpy driver tubes with high Rp was never a successful design measure in my hands.
Unfortunately I know of no octal tubes that have 40+ mu and low plate impedance.
Kevin Carter
K&K Audio
www.kandkaudio.com
![]()
My experience in retail audio taught me that most systems have too much gain, not too little, because manufacturers err on the side of more gain than less, because consumers can always reduce gain with the volume control. Unfortunately, that results in a noise and dynamics trade-off if carried too far, though.Can't agree with you more on this.
That said, the 5dB gain of K&K Raleigh linestage seems to be surprisingly low to me. I had looked at the schematic at your site a number of times. A very rough thought of a CCS loaded 6N1P should yield a mu around high 20's to low 30's. That's around a gain of 28dB to 30dB.
![]()
You are right about the gain of a 6N1P, but we are stepping the signal down 6dB in the input transformer and 18dB in the output transformer, so the ~30dB from the 6N1P becomes 5dB. Either the input or output transformer ratios can be changed for more gain readily when the kit is built.
Thanks for the information.Pardon me but can't resist wondering anything gained from stepping down a total of 24dB. Chances are this is not around current as 6N1P is a high current, high mutual conductance tube among twin triodes.
![]()
.
![]()
Hi HKL,What is gained is low gain.
The point is that most people do not need a lot of gain in their system and end up using their volume control from 7 to 8 o'clock. While developing the line stage I listened to all of the different transformer ratios with resulting gain differences. The best sounding was with the low gain. I don't know if this was because of the characteristic of the transformer ratio or because of the characteristics of the amplifier at low gain.
So what is gained is better sound.
It sounds interesting too.Mind if I ask one more question: have the tube choice, say, a lower mu one, been given a thought?
![]()
Hi HKL,Yes we did audition many tubes along the way. I started with the 5687 and tried several variants on it. Then the ECC99 and 6H30P. I liked the 6H30P, which has about half the mu as the 6N1P, and stayed with it for a while. In this circuit the 6N1P sounds marginally better than the 6H30P so I now use it. There are many factors to consider when designing a circuit; gain is only one. For me, the prime criterion is sonic quality. First get that and then do what you need to do to get your gain target, which I believe should be moderate.
Did you change any settings when you change to 6n1p.
Hi Alan,I ran the 6H30P at 20mA per side and the 6N1P at 8mA per side.
This post is made possible by the generous support of people like you and our sponsors: