• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Tube Preamp: Remedy For Digital Glare?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm thinking of building a tube preamp. Which Aikido Linestage (octal, 9-pin, tube grouping, rectification type) will have a sonic signature that's tonally neutral, extended bass and highs, lush mids, and wide and deep soundstage with holographic imaging?

After building the SE 6N1P/KT88 amp, this will be my 2nd tube project. I don't need a tube preamp greater than 8-10dB gain. My solid state amp is an Odyessey Dual Mono with unbalanced 22k inputs.
I don't necessarily need to use these tubes, but I've collected the following matched pairs over time:

6SN7GTB Sylvania (4)
6SL7GT RCA (2)
12AX7 Telefunken Smooth Plate (2)
12AX7WA Philips (2)
5963 RCA (6)
5751 EH Gold (2)
6922 Seimens Halske CCa A-phi (2)
CV4024 12AT7 Mullard Military (2)


Any thoughts and recommendations would be greatly appreciated. Thanks!
 
Hi,

I have build some Aikido preamps. Here are my experiences.
Aikido doesn´t have global feedback, so quality of the parts (active and passive) and PSU will influence sound somewhat.

Definately go with octals. 6SL7 and 6SN7 sounds much better than 12AX7, 12AU7, 6DJ8, 12AT7.

Use high quality resistors for grid stoppers and cathode resistors. I have used Dale RN65D. Difference between them and ordinary cheap metal films is big.

Run tubes with higher current and high B+ voltage as possible (340V limit for octals, because of cathode to heater voltage limit).

Regulate heaters and B+ supply. Regulated DC sounds cleaner with better bass than AC and they eliminate hum from filaments. Use fast-SOFT recovery diodes in PSU.

For your situation I recommend you all octal 6SN7 aikido.
 
I'm thinking of building a tube preamp. Which Aikido Linestage (octal, 9-pin, tube grouping, rectification type) will have a sonic signature that's tonally neutral, extended bass and highs, lush mids, and wide and deep soundstage with holographic imaging?

If you build the circuit properly, use a reasonably high B+, and keep the power supply noise reasonably low (the noise cancellation is good not great), the Aikido linestage will have no sound whatever. Your desire for "tonally neutral, extended bass and highs, lush mids, and wide and deep soundstage with holographic imaging" will be a function of the recording, the power amp, the speakers, and the room. You can't really ask more from a preamp than that. I've built three different versions and they all work very well, irrespective of the tube choices (which are a function of the gain needed).

6SN7 will get you roughly 20dB of gain, which may be a bit high. The other tubes on your list will give even higher gain. For your 8-10dB target, you need to look at lower mu tubes (e.g., 12B4, 6BX7).
 
I have never quite understood the attraction of the Aikedo topology. It uses an extra tube to provide some power supply rejection that could more easily be acieved in a pre amp with a few RC sections. If you want transparency then, as others have said, use the 6SN7 with its low intrinsic distortion. Use it in a mu follower or with a CCS plate load.

Cheers

Ian
 
I'm thinking of building a tube preamp. Which Aikido Linestage (octal, 9-pin, tube grouping, rectification type) will have a sonic signature that's tonally neutral, extended bass and highs, lush mids, and wide and deep soundstage with holographic imaging? ... My solid state amp is an Odyessey Dual Mono with unbalanced 22k inputs.

22k...what's the sensitivity? You might find that the hiss of the tubes is far more audible with an amp that has an impedance of 20k!
 
And you still can't hear the difference between two coupling caps, for example?

Only if I peek, or if one is defective, or there's a serious difference in the frequency response. Just like anyone else with normal-to-good hearing. See, for example, the Bastard Box section of "Testing, One, Two, Three...".

http://www.linearaudio.net/images/LA Vol 2 Yaniger(1).pdf

In any case, this is seriously off topic. Let's stick to helping the OP select the correct tubes to hit his gain target, shall we?
 
Thank you all for the sage advice! There are some heavy hitters on tubes on this site, and I am glad for the useful responses. It seems that the 6SN7 at ~320V B+ is the best tube to use. I will also experiment with PIO, Al foil caps for best sound.

Correction: My Odyssey Dual Mono amp inputs are actually 18k Ohm, 1.0V sensitivity, 34X voltage gain, and 31dB power gain. Will I be OK with an Aikido LS? What's the output impedance of the Aikido Linestage?

My speakers are DIY Scanspeak 21W8555/8545/9800 with series fed XO and Mundorf Supremes. It gives excellent spatial imaging, air, transparency, and clarity. The top and bass cabs are bi-amped.
 

Attachments

  • image.jpg
    image.jpg
    289 KB · Views: 464
It seems that the 6SN7 at ~320V B+ is the best tube to use.

If you can go higher on B+, that would be even better. It will drive 18k loads without even breaking a sweat. Output Z will be somewhere south of 1k.

The gain of the linestage using 6SN7 is still going to be way too high; if your power amp really has a 34x voltage gain (31dB), you really don't even need a preamp, a simple volume control (with perhaps a unity gain buffer) will be more than sufficient.
 
To be honest, if that is the qualities that you require, and you already have an amp with a lot of gain and a lowish input impedance to boot, I would not look at the normal Aikido LS at all. One option is the Hypnotise DCB1 buffer by Salas, or, if you really want tubes, a Cathode Follower like one by Broskie (ACF-2) might work and will give you unity gain and enough drive to drive those lower impedances. I don't know, maybe using 6h30 tubes in the second stage.

But I am just a mere noob still. Somebody like SY and other have vastly more experience in these matters.
 
If you are going to get PIO caps, do yourself a favor and just get Russian PIO. They are far cheaper than the western versions and sound (to me at least) very good. You can search on this forum to find out more about them from other members.

I built a 6sn7 srpp preamp to use for testing builds. I could tell the difference between the K75 type and the K42 type. I stuck with the K42.
 
Disabled Account
Joined 2002
One option is the Hypnotise DCB1 buffer by Salas

Good hint, no heat, has a muting circuit and low output Z and sounds fantastic. No gain and no (PIO) caps ! Best sound quality is when using no caps at all.

Why would one use a preamp with high gain and high output Z to a power amp that also has a high gain ? IMO it will only color the sound with no added value except for having max. volume at 9 o' clock and more noise than strictly needed.
 
Last edited:
I'll throw my .02 in for what its worth. In my opinion after building a few of these is the octals are my favorite pick. I like the 6SN7 design with a lot of B+. Secondly still having some hearing left I cannot tell the difference between different resistor brands/types and or coupling caps. There are those that may agree or disagree with what I am saying and to each his own. To those that don't see what the fuss is about I challenge you to build one and judge for yourself.
 
I'd use one triode of a 6SN7 for gain, and the other as a follower output buffer. I'd resistively attenuate the signal after the first triode if needed, which will also reduce noise from the first triode stage. Without a current source plate load on the first triode, the circuit will be sensitive to power supply hum and noise, but a multi-section RC filter may be enough there. The filiment supply should be elevated by about 90 volts DC, relative to circuit ground (an RC filtered divider off the high voltage), so you don't exceed the filament to cathode max voltage spec in the follower stage. Put a passive Rf filter at the input (1kohm R in series and a 1nF polypropylene cap to Gnd for rolloff above 160kHZ.

Personally, I'd have this circuit drive a James style tone control circuit, which would then be followed by exactly the same circuit again to make up for the gain loss. Or you could sell out to the myth that tone controls are undesirable.
 
Disabled Account
Joined 2002
So, one Aikido with too much gain , attenuate the amplified signal, then a tone control, then an Aikido again with too much gain. All connected to a power amp with pretty much gain.... You tube guys are funny. No wonder differences are heard ;)

If you have too much gain you can skip the whole device (or add a passive undesirable tone control). All together it's like adding a fifth wheel to a fine running 4 wheel car.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.