12AX7 ECC83, le grand test.

Merci pour les infos

J'irais dans mon magasin de musique pour voir si il ne sait pas avoir d'autres marques / revendeurs.

Sinon pour les spec. de cette lampe en particulier, personne n'a jamais testé?

Et si on pouvait donner un conseil, perso je dirais : n'achetez jamais les tubes chez les revendeurs qui pratiquent des tarifs défiant toute concurrence ... Vous aurez systématiquement des surprises !! Tester correctement un tube prend du temps, et donc coute de l'argent (d'autant plus que les revendeurs les + sérieux ne mettent en vente que les tubes les meilleurs, et ils ont bcp de déchet, et je doute que les fournisseurs fassent des reprises à taux plein sur les tubes qui ne passent pas les tests ...)

Et quand vous voyez un revendeur qui propose 3 tarifs différents sur une même référence de tube, fuyez ... (au moins le tarif le + bas !!)

Bonne fin de soirée, à+

En consultant ton site, je vois que tu vends des tubes (retubage ampli dans les tarifs). Je ne t'accuse de rien, mais en temps que vendeur, tu vends tes produits et c'est normal.

ELECTRO HARMONIX ECC83-Gold testing

The one thing I agreed on with EHX is, that I have much higher standard for testing the ECC83 as they have.

The EHX ECC83 is a very good HiFi tube, but unfortunately test results of those are often very disappointing. I have been in tubes professionally since 1995, and let's just say I know a little something about tubes. Also I think I have invested pretty well in good test equipment, and I happen to know a little something about tube testers also. So here I am trying to do the best I can.

I received in 2006 and 2007 several batches in 100pcs quantities of the EHX 12AX7-Gold, directly from the factory, that do not pass my incoming inspection. Please note, the factory in Russia ships those out untested, and EHX does not offer factory tested tubes of 12AX7. (though I would LOVE TO buy those with some specified form of factory testing)

I have a degree in power electronics, and I have no need for hocus pocus just to make the story interesting. An Electron tube is an electronic device, and all methods about how to test tubes have been documented already before I was born.

So nobody is in need for new test methods by tube doctors in the internet. For me, a tube test specification is something that is printed somewhere, and then anyone can repeat that, with instruments of his choice.

To prevent discussions, the manufacturers specifications are what you need. So either a manufacturer makes his own data sheet of his own ECC83, or if they don't have a data sheet, we can use the data sheets of RCA or Philips.

Not so with EHX. No specification of how to test their tubes exists. You can do what you want, but if you send back too many tubes that you say are bad, and they say they are good, you have to pay 25% re-stocking fee.

That works fine, as long as most of the tubes of a batch are accepted. It gives problems when I think most of them are rejects. With some shipments I can use 95% and they do replace the other 5%. Currently I have a shipment of 200pcs that was a replacement already and again 60% is nor passing my tests.

So do I test unreasonable? I don't think so, but the problem is EHX refuses to tell me how I must test them. So I test them as written in the FUNKE handbook, and in other literature, saying they must draw minimum 70% of plate current. Honestly, would YOU want to buy tubes with 60% plate current. Still for EHX I was told that is no reason to reject them, as long as they play good. For that they use a KAYE Audio labs tester, which basically test the tubes for gain and noise. However, GAIN is no good criteria for incoming inspection.

So we have to continue to select those EHX 12AX7 the regular way, as 50 years ago. That is standard parametrical testing. Nothing more, nothing less.

Why this is written here? It is no complaint about EHX. They are a good company who do what they say and say what they do. But what they do is sell untested tubes. Nice for guitar amps, but not for HiFi. I was told 99% of the buyers accept those tubes "as is". Well I am amongst that 1% then, but it's where I am going to stay.

So please understand, the SELECTED versions of 12AX7 are quite expensive, and you receive the few ones that really pass our test methods. So, selecting for 12AX7 can not be done at low cost. The cost factor is that when we buy 200 tubes, only 70 or so can be used as prime quality, 70 are second choice but fair enough to sell just above cost price, and another 60 we can send back and must pay the restocking fee and shipment. So the profit for 200 tubes has to come from the 70 good ones. These are the ones we offer ad 12AX7-GB (GB=Gold-Balanced).

July 2008. This information I have received after many phone calls with Electro Harmonix. All we wanted were engineer's guidelines how to test this tube, but we didn't get it. The final answer came from Chief Engineer, JC Morrison. (text not exactly quoted, but the content is) "On a parametric tube tester, you will not see the same test results for this tube, as you would get from NOS tubes. The Electro Harmonic 12AX7-EH needs a different grid voltage, in order to draw the same current as NOS tubes . If you observe this, it is normal. This was done to achieve the EHX typical sound for this tube. It was the result of a lot of research. Test settings are subject to change, and are not published. The tubes are guaranteed to be good, but parametric testing for tube for HiFi purposes must be done by the tube dealers. There are no factory guidelines for this, each tube dealer must do it his own way"

Now, this was not a satisfactory answer. It may be right for a guitar amplifier, but for HiFi it is not, and we want tighter results. So what we do at Jac Music, we collected statistical data of about 500 of those tubes we tested, and determined the average grid voltage for making the plate current the same as with NOS tubes. Plate current is for this tube by definition: 0.84mA.... 1.71mA, by old Philips data. It appeared a negative grid voltage of 1.9V is needed to achieve this. (and not 2.0V as written in historical data sheets). This makes it clearer once again how important tube matching is. Pairs we ship are like: 0.9/0.9mA or 0.9/1.0mA or 1.0/1.0mA etc until the end of the range: 1,7/1,7mA. So the matching guarantees you 0.1mA maximum difference, which is really tight, and can be done with only the best tube testers. So that 0,1mA maximum difference is within one bulb if you buy one balanced tube, and for all four triodes too, if you buy a pair. Inquire for quads if you need, it is often no problem.

June 2009. Another batch of bad tubes. 1/3 of those do not pass incoming inspection, by normal test methods. . My sales person offers me to send back the bad tubes, and if they are found to be good after all I need to pay 25% restocking fee.

So here we are, with a very unsatisfying situation. EHX refuses to tell me how I must test their tubes, the only thing they come up with is a tiny little tester from KAYE AUDIOLABS, that has only a good/bad scale (for "gain" as I learned) . This tester is their reference. I looked up in the internet what this tester is doing, it tests only Gain and noise. Now that is too simple for me. Sure I can test gain also, and gain is fine with almost any tubes EHX sends out. Besides on my L3-3 and Amplitrex, I can test gain numerical, and not need a silly "Good-bad" scale. HOWEVER it is not that simple in real life to test a tube. I would say if you know how to test a tube really well, you will agree with me, and if you do not know well, there is no point in trying to explain quickly here what needs half a book.

Donc je re-précises la chose : Je ne t'accusais de rien, mais je voulais que tu précises ton propos.

D'un autre côté précédemment, j'ai acheté les fameuses Electro Harmonix chez un autre vendeur, je les ai installés et ca sonnait très mal, pas de volume, aucun gain, des grosses m... payées plus chères que les JJ chez Sven. J'étais pas content.

Plus sérieusement, ce qui me permet de juger de la qualité d'un tue, c'est un ami qui s'en charge pour moi, un homme de 79 ans, qui fut entre autres un des concepteurs de l'électronique à lampes du Revox F36 puis du Nagra, ami de Fred Kaftanski et je penses de Jean Hiraga. Au vu de mes connaissances limitées dans la technique des tubes (je suis un enfant du transistor, le tube n'est pas ma génération) si cet homme me dit "ils sont valables" après le test au lampemètre, je considère que c'est du bon.

Je m'en remets à l'avis des spécialistes.

Par contre une remarque, le µ n'est pas le gain, et j'avais lu que s'il y a écart avec la norme, c'est aussi parfois (pour certaines marques anciennes) un choix de conception de limiter l'émissivité et d'augmenter la résistance d'anode en diminuant le µ pour au final un gain externe en tension identique (ou supérieur)
Il faudrait pouvoir mesurer la "résistance de plaque" de chaque triode pour s'en assurer.