Schiit Lyr Tube Rollers

[Astral Abyss]

Ok...

After I got the tube imp I realized that most of the auctions I bought on ebay are not testing nearly as strong as advertised.

It was sort of a testing and crying kind of thing I still cannot judge how much of a difference tube strength does to sound but I guess this will come with time.

What tester did you buy? Testers are normally pretty pricy aren't they? I'm also guessing if they aren't calibrated they're pretty much worthless?

 

[kolkoo]

It's called the "Tube Imp" it's a sort of a mini modern digital tube tester.
 
http://www.britishaudio.co.uk/tubeimp.html
 
It has some limitations like max 12 mA on HT current which can be overcome by testing let's say the 6N23P at Ua=90V and Ug=-1.5V which then makes 100% Ia = 11mA and 100% Gm = 11.5 mA/V
 
but it does a great job of matching, it measures cathode emission, transconductance and Gm.
 
So far I have no reason to doubt it as it measures the Lyr 2 stock tubes at 100% of their specs and I've got some nicely matched 6N23P from a good seller that also shows the tubes that I got perfectly matched and really strong.
 
+ I've tested a batch of 10 completely identical HGs made in the same month III 1975. And I've seen how the strong ones measure etc.
 
Ofc the possibility exists that it's not accurate but so far it seems good.

 

[MWSVette]

What tester did you buy? Testers are normally pretty pricy aren't they? I'm also guessing if they aren't calibrated they're pretty much worthless?

 
I bought a Lafayette TE-15 on ebay for $50.00.  Cool little portable unit.  Tests for shorts and quality of emission on each triode of the tube. 0 to 100.  50 being good 100 new.  Look on ebay there are always quite a few different kinds listed.
 
As for being calibrated even if the measurements are not exactly accurate.  If the unit shows the same measurement on both triodes then they would be matched.  And the unit easily shows any bad tubes.

 

[Hardwired]

  It's called the "Tube Imp" it's a sort of a mini modern digital tube tester.
 
http://www.britishaudio.co.uk/tubeimp.html
 

 
I looked at that a few months back, but the exchange rate had it over $600 at the time. Now for roughly $350 I can take a chance on one, so I ordered it. Thanks for reminding me about the Tube Imp. My experiment with the DIY kit for a uTracer ended with a lump of worthless circuitry so maybe the Imp will work better for me since it doesn't involve a soldering iron.

 

[kolkoo]

   
I looked at that a few months back, but the exchange rate had it over $600 at the time. Now for roughly $350 I can take a chance on one, so I ordered it. Thanks for reminding me about the Tube Imp. My experiment with the DIY kit for a uTracer ended with a lump of worthless circuitry so maybe the Imp will work better for me since it doesn't involve a soldering iron.

Oh that's sad to hear
 
I intend to build a uTracer to see if my tube imp measures accurately
 
Now I'm abit scared.
 
To be honest the tube imp is great for matching.
 
Just have in mind you wont be able to test at 100% of specs.
 
But that;'s actually fine in theory if a tube measures at 100% at a bias point where the current is lower
it should measure the same or better at higher current.

 

[Hardwired]

Oh that's sad to hear

Oh, it gets sadder when you realize I bought 2 kits, knowing I would probably screw up the first one and then learn from my mistakes on the 2nd.I barely got started on the 2nd kit when the power supply doesn't work and I gave up on it too. Want a couple of barely used kits that need a little work?

 

[tvnosaint]

Kinda makes your name a bit ironic or a play on words.

 

[Hardwired]

Kinda makes your name a bit ironic or a play on words.

Ha! Didn't notice that, but you're right. My username is from the title of a cyberpunk novel but maybe I should change it to NoSolderingPlease

 

[kolkoo]

So I've been reading more about matching, tubes, how the Lyr works Downloaded some electronic tube pdfs from the 1960s etc.
 
My Tube Imp can measure cathode emission (Ik, which should be the same as Ia in preamp circuits as there is no screen and stuff like that), Gm (transconductance) and Mu (gain).
 
So I started matching my tubes by setting a fixed bias point, let's say for 6N23P around Ua = 90V, Ug=  -1.5V and calculating the average 100% from the specs which in this case is
 
100% Ia = 11mA, 100% Gm = 11,5 mA/V , Mu = 31 .
 
So I tried to get two HGs ( I have around 13 of those, 8 of which are quite good, I might end up selling two pairs of these after I'm sure my matching methods are optimal for the Lyr)
with as close as possible in percentage matching in all 3 characteristics - and they sounded quite good.
 
Now then I started reading about how lyr is biasing tubes - it seems that lyr is using a constant current source type of biasing:
Quote Jason Stoddard:
"4. Lyr, Lyr 2, Mjolnir 2, and Valhalla 2 all use constant current source biasing (rather than, say, resistor biasing), which locks down the amount of current flowing through the tube and results in an invariant operating point regardless of tubes used—no manual or automatic bias adjustment necessary."
 
So I looked more into matching and found many places where people say you should match preamp tubes by Plate Restistance Ra/Rp  which can be calculated by the formula
 
Ra = Mu/Gm.
 
One of these resources also stated the Gm and Mu (Ra) measurements should be taken at the same anode current as they quite dependent on it.
 
Then I also found this tutorial on tube matching http://www.augustica.com/Tutorials/Tubes/matching_triodes_and_pentodes.htm.
 
It seems here that on the amplitrex they have set voltage of 90V and desired plate current of 15 mA and the amplitrex found that it can achieve that at
Ug = -1.2 for one section and -1.3V for the other. Then it measured the Gm and Mu at the same plate current and it calculated the Ra - and the matching is done here by Ra.
 
So I put 90V on my tube imp and adjusted Ug in the range of -1 ot -2V so that each triode section I'm measuring is at 11 mA cathode current. I took Gm and Mu measurements and I matched a pair of two HGs with the smallest Ra differences.
 
Now what I can say the result was different. I can't say if it was better objectively obviously - but I liked it, it was definitely different.
 
I will continue trying stuff out with matching on my Tube Imp and when I'm sure I can optimally match for my Lyr I will be reselling some of the tubes that are not my favorites
 
 
EDIT: It seems that the maxi preamp (which costs a truckton) also measures Mu and Gm at a constant current http://www.maximatcher.com/maxipre_manual.pdf. And the matching can then be done by (Mu,Gm) and (Mu/Gm), I'm starting to think this is the proper way to match
 
Here as well http://www.amplitrex.com/download/AT1000_Manual_09-2009.pdf they have fixed bias and auto-bias matching. The Lyr seems to be auto-biased (probably using similar principles) so it would make sense to match tubes this way.

 

[syntheticfish]

Well its been an exciting day! I woke up to discover that my NOVIB socket Savers, 1969 Amperex Orange Globes, and 1970s Mullard CV2493s (courtesy of Nightlight) have all arrived together in the post! As Tobias Funke would have said, "let the great tube rolling experiment begin!" - if you don't know what I'm talking about, you clearly don't have a net flicks account...
 
So far I've been listening to the mullards. Now its early days, but my immediate response was shrill... I'm coming from stock Schiit tubes, but right now I'm not liking the change. I'm hoping some burn in/adaption will change that, but this is my immediate reaction... I haven't heard anyone else accuse Mullards of being shrill... I must also acknowledge that I've been listening whilst working and as it is now nearly 5am UK time, I may be missing something due to sleep deprivation...
 
On an irrelevant note, these tube monger socket savers make the amp look a bit bloody silly - even if the tubes are now much more accessible! 

 

[flyingmonkey]

Have had my Lyr 2 for a few weeks which I feed from a Pioneer XDP-100R into a Chord Mojo and my headphones are HD700, I upgraded from a Sony Nwza 15 feeding a Littledot 1+ with Mullard 4010 valves and a muses02 op amp.
 
The Lyr 2 was better even with the stock made in canada tubes but I was not convinced that they were the best, so in went a set of Tesla E88CC valves and these improved the sound better dynamics and improved bass and mids and a nice airy treble.
 
Not being able to resist in went a set of bugle boy d getter ecc88 tubes another improvement this time a wider soundstage and a better balance to me.
 
Then a pair of Tungsram E88CC military tubes better than the Tesla but in a similar way but not as good as the bugle boys.

 

[tvnosaint]

My tungrams were boring. A side grade to the JJs . So we're matsu****as. Bugle boys were pretty good. The lyr really likes miniwatts from heerlen and suresnes. I like the ecc88 Blackburn mullards in it. They lack soundstage but sound so pretty I don't mind their flatter nature.

 

[NightFlight]

  Well its been an exciting day! I woke up to discover that my NOVIB socket Savers, 1969 Amperex Orange Globes, and 1970s Mullard CV2493s (courtesy of Nightlight) have all arrived together in the post! As Tobias Funke would have said, "let the great tube rolling experiment begin!" - if you don't know what I'm talking about, you clearly don't have a net flicks account...
 
So far I've been listening to the mullards. Now its early days, but my immediate response was shrill... I'm coming from stock Schiit tubes, but right now I'm not liking the change. I'm hoping some burn in/adaption will change that, but this is my immediate reaction... I haven't heard anyone else accuse Mullards of being shrill... I must also acknowledge that I've been listening whilst working and as it is now nearly 5am UK time, I may be missing something due to sleep deprivation...
 
On an irrelevant note, these tube monger socket savers make the amp look a bit bloody silly - even if the tubes are now much more accessible! 

 
http://www.head-fi.org/t/683012/bottlehead-amplifier-discussion-comparison-thread-crack-sex-mainline/1125#post_11530125
 
Use one of those to break in shrill tubes. 

 

[syntheticfish]

   
http://www.head-fi.org/t/683012/bottlehead-amplifier-discussion-comparison-thread-crack-sex-mainline/1125#post_11530125
 
Use one of those to break in shrill tubes. 

 
That's a great solution! I might get my soldering iron out!

 

[kolkoo]

http://www.ebay.com/itm/4-x-E88CC-ECC88-6DJ8-6922-6N23P-NEW-1970s-LOOK-BIG-LOGO-REFLEKTOR-/172074172027?hash=item28106b7e7b:g:n44AAOSwqYBWpAWW
 
HGs boys 6$ a pop. I got some and will match them with my tester.
 
I already have 3 really good pairs + 1 good pair.  So if I get a few more out of those I might begin selling them to headfiers if anyone is interested ofc.
 
Wouldn't recommend buying these without a tester though - as they can be really bad and even if there are some good ones to match - you don't know it.
 
 
EDIT: Knew it was too great to be true - I received the following message from the seller approximately 1 day after dispatch:
 

Just little thing.
We send for you 1970, 1971,1974 years. Selected quads.
Don't worry all with BIG logo.​

 
"just little thing" oh well...