Tube rolling thread | UltraSonic Studios
Mar 27, 2024 at 7:53 AM Post #3,481 of 3,640
@Deyan is building great adapter, nice looking, top quality and you can disuss with him in HeadFi to optimize tube/socket usage. I've also used in the past chinese adapters and got some faulty ones specially small one using adapter board and crappy socket... I'm now building mine when i have time and get good base/socket in China or from old rusian stock for 7/9 pin socket or GU50 ceramic ones. (Final result is audibly OK but much more ugly than Deyan :frowning2: ).
If you have in your device not so good socket, using adapter and/or rolling tube too much can lead to loose connections. Jut pay attention from time to time that each pin is individually secure and if not use a micro screewdriver to inch a little the connectors.
How hard would it be to rebase the tube? I’ve seen places that do that with loctal tubes. If you could turn the side contact tubes into octals that would solve a lot of problems.
 
Mar 27, 2024 at 8:00 AM Post #3,482 of 3,640
4V tubes. Very nice.

If I understand correctly, this is the B5 socket:



This is something totally different than the P8A side contact I use. I have no idea about adapters for these.
IMG_0035.jpeg


Something even more freakier!

Supposedly those side terminals are meant to be tapped into, but I have never seen anyone use them.

I was told if not done properly, a poor connection will cause damage to the amp so I just keep them as a display piece

In Europe not, but after reading this I think I might give a try with 1101 Audio. I would incorporate some of those ideas in a build with P8A sockets for 6.3V/4V old european pentodes. I think it will be something revolutionary. I find interesting the concept of "servo-biasing".

I will explain why. During my experiments I realized that some tubes do not perform optimally at a fixed anode voltage. Usually tube amp builders choose a decently low value for the plate voltage which will let you use safely mostly all tube types. This does not mean you will optimally use those. I would like to try some of these old pentodes at different voltages, for example tubular newer versions which allow higher plate voltage or currents to bias higher while older pentodes with more restricted ratings to bias safely lower. In this way I think we could achieve the "sweet spot" for each of them and use them properly.

I was lucky to obtain so good results here with 4654 and the rest of the family but I am sure I can do more.

I am on my final stretch of waiting for my amp from Mischa, I believe given ample time he is technical enough to make what you want!
 
Mar 27, 2024 at 8:07 AM Post #3,484 of 3,640
What have you ordered from him?

I don't think he posted it yet - just in case he wants to do some marketing I won't post much about it, but its something I sold off my excess amps and headphones for :)

Mine would be a customized overkill version of it, but it goes a different direction than what you have in mind (pentode focused?), mine will be triode-based
 
Mar 27, 2024 at 8:09 AM Post #3,485 of 3,640
Mine would be a customized overkill version of it, but it goes a different direction than what you have in mind (pentode focused?), mine will be triode-based
These pentodes are triode strapped. But I can see if ultralinear will make sense. It's not a finalized idea.
 
Mar 27, 2024 at 8:15 AM Post #3,486 of 3,640
These pentodes are triode strapped. But I can see if ultralinear will make sense. It's not a finalized idea.
Most excellent, I did spend quite a long time looking for triode curves and datasheets while waiting for my turn to pay the deposit :p

It was a fun journey. Maybe one day we'll have a headfi utracer guy around here to help us
 
Mar 27, 2024 at 10:22 AM Post #3,487 of 3,640
How hard would it be to rebase the tube? I’ve seen places that do that with loctal tubes. If you could turn the side contact tubes into octals that would solve a lot of problems.
On Euphoria thread yeras ago, some folks did exactly what you suggest ! Remove the base and change to the target pinout. This is fine if you do not want to sell anymore your tubes, and are not a tube collector at all... Beside this you can in most case a way to rebase most of your tubes... You also have some special glue for hot temp to use if you want to have a solid base and not a loose one.
It was a fun journey. Maybe one day we'll have a headfi utracer guy around here to help us
I can post some of the curves I do if you want, but not sure we have all the interest in such curves...
But just to get simple measurement, duovac do the job very well.
 
Mar 27, 2024 at 10:25 PM Post #3,488 of 3,640
Went back to the 14n7 in the preamp. Sounds really good now. I’m going to try the el38 with cv4062 inputs. The el38 is the highest gain output tube I have I think. After that I will try the kt81 on input duties. Anyone here ever try kt61 as an input?
 
Mar 27, 2024 at 11:11 PM Post #3,489 of 3,640
Went back to the 14n7 in the preamp. Sounds really good now. I’m going to try the el38 with cv4062 inputs. The el38 is the highest gain output tube I have I think. After that I will try the kt81 on input duties. Anyone here ever try kt61 as an input?
I did not made any try with KTxx in input but this one seems to be like a double 6M6 (twice more power) and 6M6 is working as great as EL11... in input and output in our eternity amp. (In Elise and Euforia, the plate voltage (140V) is a bit low and this introduce a bit of pleasant distortion).

KT61/KT81 from spec is for me a bit on the boundaries for input/driver not too linear at low power. I do not see from the spec amplification factor, and do not have any to get curves and spec at usage point.
This should not blow your amp anyway if the plate voltage is not that high and grid low enough...
 
Mar 28, 2024 at 12:01 AM Post #3,490 of 3,640
I wonder if running my synth into my amp directly would cause any trouble? The signals coming out of them can be quite hot, up to 10v. When I told Tomas that my preamp has lots of drive he asked if I wanted to bypass the input/driver tubes. Had never heard of an amp that did that but if the signal is hot enough and has enough drive I guess it could work.
 
Mar 28, 2024 at 6:22 AM Post #3,491 of 3,640
I wonder if running my synth into my amp directly would cause any trouble? The signals coming out of them can be quite hot, up to 10v. When I told Tomas that my preamp has lots of drive he asked if I wanted to bypass the input/driver tubes. Had never heard of an amp that did that but if the signal is hot enough and has enough drive I guess it could work.
10V -5/+5 in XLR output of your synth in open circuit or feeding input impedance of (pre-)amp ?

With Synth, you can generate a Sine at 50 or 60Hz and use a voltmeter in AC mode directly plugged in the amp and measure inside the plug, you just have to use a std pro cable XLR 2 XLR cable and remove in one side the plug housing.... You will get a rms voltage value that should be multiplied by Square 2 to get peak value but this is this RMS value the most relevant for the subject.
With this you can get a better idea of what will float to your amp.... Up to -3/+3 is still OK i many pro devices. (I have a dac producing this value, Lampizator folks also have hot value out of the DAC...)
As you have an input pot just startup at 0 should also be safe in most case.

Bypassing input tube is in general not a very good idea as to feed output tube you must reach some critical values unless you have normalization circuit. The opposite, bypassing output is current to use only the "pre-amp" section with small adaptation....
 
Mar 28, 2024 at 12:08 PM Post #3,492 of 3,640
I wonder if running my synth into my amp directly would cause any trouble? The signals coming out of them can be quite hot, up to 10v. When I told Tomas that my preamp has lots of drive he asked if I wanted to bypass the input/driver tubes. Had never heard of an amp that did that but if the signal is hot enough and has enough drive I guess it could work.
I’m not sure of what synth you have but it should work- signal levels/master volume output should be set very low to start with and then increased until you’ve found a balance between your listening level (amp) and how the input (synth) is working so you’re not clipping. I used to do this all the time with various rack effects that all seemed to have slightly different output levels. Keep your eye on the internal/stored program levels though as they may vary as well.
 
Mar 29, 2024 at 12:27 AM Post #3,493 of 3,640
I had planned on trying el38 output tubes in order to get away from the closed in sound I was getting with cv4062 inputs. While digging up the el38 I came upon the vt60a. Maybe it wasn’t a matter of gain but I just needed a more open sounding tube? The vt60 has always sounded that way so I figured it was worth a shot.

Sigh, no I was having the same issue. The sound was uncharacteristically subdued and strained. On a whim I decided to try another input tube using the same adapter. I put in the CSF 6005 tubes with appropriate adapters. Whoosh, everything has opened way up. I still have the great bass but now the top end has come back as well as that effortlessness I associate with the vt60.

The different sound shouldn’t be too much of a surprise really. These are very different tubes. The cv4062 is supposedly in the same family as the el84. A great output tube to be sure but I have never heard of them being used as an input or driver tube. The obvious answer as to why is that they just aren’t very good at it. There are plenty of designs in the DIY world using triode 6v6/6aq5/el90 as a preamp tube at least. My guess is that the cv4062 as well as the el84 present too difficult a load for a preamp or DAC to drive them properly. They just require too much oomph or something to work well in that position. It makes me worry about the kt81. It is the same tube as the kt61. The kt63 is similar to a 6f6 and RCA published data about using a triode strapped 6f6 as a driver so I’m sure those will sound good. I have several flavors of 6f6 that I will try later. I don’t know how the kt61/kt81 will respond. Only one way to find out lol. I’ll keep these in for the time being and then try the Brimar, Mullard, and Telefunken 6aq5 tubes as well.
 
Mar 29, 2024 at 2:35 AM Post #3,494 of 3,640
I don’t know how the kt61/kt81 will respond.
I tried some GEC KT61 here on input and sounded OK (I know this does not say much but I do not remember exactly how, it was just fine form my point of view). If I remember correctly it was with KT66 grey glass on output. I have a quad of KT61 here somewhere, in the storage. From what I have experienced here, pentodes in input strapped as triodes tend to sound more open. I am unsure why, I just can say that KT61, EL38, EBL1, EL3 have all something in common when used in input. But the problem of these is usually the low gain, I would not go under 20 in input because it just does not sound right on my amp.
 
Last edited:
Mar 29, 2024 at 7:48 AM Post #3,495 of 3,640
If you love so much the british sound and pentodes, I'll come to you with a triode. LOL Yes, some triodes can be used in Eternity in output if I can bias them with less that -54V. Not really sure it makes sense to use this kind of tubes but why not. Let's see how this goes.

So let's look at GEC A2293, this one was used by others on different threads, usually in input, but for me it's clear an output tube.

GEC_A2293_pair.jpg


I have a nice NOS NIB D-getter pair here, some details:

GEC_A2293.jpg


VB = February 1964 if my GEC date codes tables are right.

Unfortunately this pair is very unmatched. Anyway, no audible difference and I can use it with some caution. I'll show some measurements:

1711712149844.png


Now, on the amp things stay an idea better (there are always small differences between measurements and amplifier). One sits at 23mA while the other at 16mA. No audible difference, so all these guys spending in vain for matched pairs are, of course, wrong. I said this many times, but well, they feel better using matched pairs, vendors make some money and all are happy. I am also happy, I get the untested/unmatched more easily and cheap. I thank them, I have more money for beer.

Here is the stronger one on DuoVac, -44V for 20mA, but to keep both in line I put around -44.4V on the amp. I prefer to have one tube biased lower than to go too high with the other.

GEC_A2293_DuoVac.jpg


And voila, on the amp with a Philips Herleen 12AT7 put there especially to remove some of the GEC flavour while keeping a decent level of bass. The british flavour is too much for me lately. Sounds sweet, nice and mid forward but it's veiled. A kind of GEC flavoured EL84 sound. Musical with this kind of british sound all love.

GEC_A2293_Philips_12AT7_Herleen.jpg


And now we reach a nice milesone, this is as far as I remember my first triode on the output in Eternity. Now, according to some GEC flyer because I could not find a datasheet for these, max Va = 300V and max Ik = 120mA, so this is a power horse and can work on more powerful amplifiers. It has 15W so ... I am a lot below.
 

Users who are viewing this thread

  • Back
    Top