What Do Tube Testers Really Bring To The Table?

[MylesBAstor]

A topic I've often pondered. Years ago, we were told the reason tube testers weren't a reliable index was that they needed to be calibrated. Then tubes needed to be tested at normal operating voltages. Then tested in the actual circuit to ensure circuit meets specs. Yet despite this, I find the results from tube testers are disappointing. And certainly no predictor of sound. Or longetivity. (Other than manufacturers and someone like a Kevin Deal, I wonder how many tube retailers do a tempered burn-in?) I won't mention the name but years ago, this measurements oriented person-you know the type, tubes have no sound but it's how they are used in the circuit- was selling these after market upgrade kits for various tube preamps and amps. Tubes measured perfectly but they sounded like expensive crap.

Are tube testers perhaps of more utility when it comes to output compared to small signal tubes? There's no question on a tube tester when you have a bad output tube. Yet small signal tubes can test good yet you know they are shot because they have no life e.g. Dynamics. Just boring sounding. Without question, tube testers have a role in matching tubes, matching triode sections or detecting microphonics. But bad or shot small signal tubes?

Has anyone found a tube tester that really shows when a small signal tube is shot?

[kcin]

You make a good point.

Most, if not all tube testers, are of little value for power type tubes as we can't run the power dissipation through the tube for an accurate analysis of its performance at different bias set points. We get relative go ; no - go readings and shorts ( a good thing). The Amplitrex supposedly solves the working voltage operating point problem.. very expensive but maybe the only alternative in today's world..

I can't really explain why certain tubes sound differently than others , I can postulate, but they do. (Capacitors make more sense to me from a material science point of view) For small signal tubes we can test for noise and microphonics and gm then experiment for tonality and extension etc. Best tester for small signal tubes is the circuit.

Outside of that, this one works for me on the noise microphonics as well as relative quality on gm. This VTV tester is no longer available but was distributed by VTV and designed by Eric Barbour of Svetlana fame. It works the best for balance , noise and gm quality compared to any other tube tester I have for small signal tubes- I have not tried or even seen an Amplitrex. As to how a particular tube will react in your circuit; well that's why we roll tubes isn't it?

[MylesBAstor]

Cool, never saw Eric's tester before.

[Dr.Ears]

Most tube testers will recognize a completely dead tube, but none have great tests for microphonics.
The problem with the Amplitrex is it will find something wrong with just about every tube, which you test, although they may sound just fine in a circuit.
There is no substitute for the integrity of the guy, who you are buying from.

[Rust]

I remember that YBA, a French company, tested transistors before they were potted/cased as a cost saving measure. This was to identify those that most closely matched their tighter specification without incurring full cost. Or so they said.

Variances in manufacturing could easily explain variances in performance, even in tubes sequentially produced on the same assembly line. A tiny bit of variance in the coating of the elements, a small variance in vacuum levels, a small variance in the geometry of internal elements. Pretty much the only way to determine such variances would be to test for the individual tubes trans-conductance curves at full voltage. I am not aware of any commercially available tube tester with such capability, such equipment is normally found in a production facility and even then tubes are not individually tested to such an extent. To time consuming to burn in and test each individual tube to such a degree.

Manufacturing becomes more difficult the smaller the item produced. Say that the achievable dimensional accuracy is 0.05" for a 1" item. That is 5%. Take the same 0.05" for a 0.25" item, that is 20%. So small tubes would be more affected by the same variances which would not be as detrimental to performance in a large tube. Consider accurate positioning of the anode, cathode and grids, plating thickness, surface area of the elements, shape of the elements and any other influence on performance.

So yes, I'd have to say that small tubes that more closely align with target specification are a smaller percentage of current production. And since they tend be utilized with small signals, it's that percent error thing compunded. Back in the days prior to transistors, tube manufacture was an entire worldwide industry and enormous resources were applied to the development and manufacture of tubes. Today it is a boutique industry. Imagine what modern tubes would be like if the modern manufacturing prowess of say an Intel were applied. Thus the cachet of NOS tubes, manufactured at the height of the tube industry, when tubes were the only show in town.

[MylesBAstor]

You know tube companies rejected tons of tubes from the assembly lines before shipping too? 😒 Somewhere on the board, I posted a picture of a landfill sized pile of rejected tubes at the old amperex factory in LI.

Pretty sure Amplitrex and old curve tracer did what you mention. But still it's how the circuit performs with the tube in it that counts the most. Which companies like cj and ARC do and what you in extra pay extra for when you buy from the manufacturer.

[Rust]

Myles - The picture was amusing. Wastage in modern production of many modern precision components isn't as bad simply because of advanced manufacturing technology, should a manufacturer chose to avail themselves of such.

I imagine that Conrad Johnson cherry picks the tubes they install in their products and carefully matches them. At that price point doing less would be unconscionable.

[MylesBAstor]

I don't think the cj is unique in this respect but depending upon where the tube is used in the circuit eg. phono vs. line stage, the rejection rate might be 50% or higher. Years ago, I saw an ad on Audiogon from someone selling some 12AU7s that they had obtained from cj. All I could think is that these tubes were cj rejects.

Sad note. Years ago, manufacturers sent the reject tubes back to the distributor/wholesaler. All the wholesaler did was put the tubes back on the shelves. So in the case say of the old GE6550s, Richardson Electronics kept putting the pentodes back on the shelves and at the end, all that were left were tubes that had been consistently rejected. But they were still trying to foist these tubes off on customers.

[MylesBAstor]

Most tube testers will recognize a completely dead tube, but none have great tests for microphonics.
The problem with the Amplitrex is it will find something wrong with just about every tube, which you test, although they may sound just fine in a circuit.
There is no substitute for the integrity of the guy, who you are buying from.

Did you ever play with George Kaye's long OOP tube tester? I though it was pretty good for picking up microphonics. Set up in a class A circuit and you could gently tap the tube and hear ringing through a small speaker or headphone jack. Like the springs on a bed.

[kcin]

Did you ever play with George Kaye's long OOP tube tester? I though it was pretty good for picking up microphonics. Set up in a class A circuit and you could gently tap the tube and hear ringing through a small speaker or headphone jack. Like the springs on a bed.

The VTV tester noted above has a similar feature.

[Dr.Ears]

I am not familar with either tube tester mentioned, but if OEM manufacturers are suppling tubes, they are your best bet.

[MylesBAstor]

Agreed and some may warranty the tubes for three or so months!

[Rust]

Shipping of tubes would worry me to some extent. Regardless of carrier, packages get treated like the old Samsonite commercial where a suitcase was given to an adult silverback gorilla

A few years back I ordered something that amounted to a high alloy steel rod. It arrived bent with the very heavy shipping tube destroyed. I tried to bend it further to see just how much force it would have taken, and it was a lot, a very surprising amount of a lot, an amazing amount of a lot.

[david k]

I have bought a few of them in the past this is my conclusion;

- None of them can measure longevity.
- Detecting microphonic tubes isn't always dependable and some of them can become microphonic soon after.
- Depending on the tube sometimes the numbers mean nothing without knowing the target parameters of the circuitry.
- Very limited value today.

david

[MylesBAstor]

Thanks for sharing your experiences David!

[Barry]

I have measured a few hundred tubes and own an Amplitrex AT1000. Some thoughts:

1) If you don't know how tubes operate, there's no way you'll understand measurements.
2) If you have limited knowledge of circuit parameters, you might still be able to test tubes if the tester operates at real-world voltages (e.g. you can do curve tracing with the Amplitrex).
3) If you know something about 1 & 2 you can tell whether a tube is being run hard in a circuit or just "loafing" and hence can learn something about its longevity in that specific application.
4) You can tell how "used" a tube is if you know what the measurements on it were when new and remeasure it later.
5) I for one would like to know mu or gain of small signal tubes especially where they need to be matched, e.g. left and right channels, and a tube tester can tell you that.
6) Using an Amplitrex you can easily verify what you are paying for when you buy tubes from such vendors as ARC. You can also see when tubes are not really well matched. Again, it helps to know something about the tubes.
7) If tubes match at one voltage bias setting and have the same Gm you can see they will match at other bias settings.
8) Again if you know the type of tube, you can tell if it doesn't measure strong, has problematic leakage, or doesn't have a good vacuum before it damages your equipment. "Green" or "Red" is a joke.
8) I'm not sure I really need to a tube tester to measure microphonics, it's pretty audible, but you can test for it.

In summary, like anything else, a tube tester is of limited value if you're not prepared to learn a few things first. The more you use one and learn about particular types of tubes; the more useful it is. I have found the Amplitrex useful in learning what little I know about how tubes operate and I appreciate that knowledge, but I understand others may not find it worth the effort. Some people really just want to listen to music....

[MylesBAstor]

Thank you Barry! I think you raised a point I forgot about! Longetivity and how hard does the designer push the tubes. Remember the Audible Illusions preamps? They ate tubes like they were going out of style. Also remember the old ARC SP10 tended to eat tubes. For some reason, Tele ECC83s never seemed to last in my old cj TEA1bc phonostage either.

[Dr.Ears]

I bought seven testers before I found one that I am satisfied with, it was pure luck.
I found a Western Electric tester made by Hickok, which actually works as it is supposed to.
Just for kicks I checked the inventory of Internet sellers of NOS tubes that I used to do business with and found they have virtually no usable inventory left.
It has been many years since I needed a tube to actually use in my system, I buy my favorites, when I see them as investments.
Most of the old Internet sellers got into tubes because of their HAM radio hobby. Even the nicest of these guys have turned into grouchy old men.
eBay essentially has put these guys out of business because they are no longer needed as intermediaries.
Stores like "The Tube Store" and "Tube Depot" are just surviving selling current production tubes. They have had to create very complicated devices to match and balance tubes and they tell me that the Russian and Chinese producers really do not accept many returns, so they have to eat the bad ones and raise their prices to account for the losses. My tube tech used to be an independent repair tech for about a half dozen US manufacturers. He told me he once got the job of getting GE's line, which produced 211's back up to specs. He finished his career as the Chairman of AT&T's personal technician. According to him GE did nothing to reject bad tubes, but did take bad ones back. I have over 1,000 of his tubes, which he simply abandoned, I have not heard from the guy in over a year and cannot find him. I have not tested many of his tubes even though about 150 of them would be marketable, If I could find a safe place to sell them. Most of the legitimate dealers, who I know, no longer accept PayPal after being burned too many times, which eliminates eBay and Audiogon. I now only accept USPS money orders and personal checks from my long time customers, but I no longer solicit them. When I have tubes, which they need I will sell to them. I am in buy and hold mode.

[Alrainbow]

Ps vane tubes look right but don't sound right. But acceptable in some ways.
I own a pair of the metal base 101 d and a real nos black base with the skinny pins I use with adaptors.
While 80 year old tubes do sound better it's not night and day. Why don't anyone make real new nos is there no demand for them. He picture I posted of the tak tubes seems to be very close to the nos
have you tried any of them.??
Also have you heard any elrogs 300B

[Dr.Ears]

I bought the Shuguang Western Electric Replicas of the 845 and the 211 over the Psvane Replicas also made by Shuguang because they have very close to the same contruction as the United Electronic 845 W with the large hole at the top, which was an efficient way of removing gases. My amp will accept either tube with a small bias adjustment. I loved the 211 in my amp until I bought a pair of very rare NOS RCA 211 black plates. I paid $1,700, which is the highest price ever paid on eBay for a pair of 211's, the previous record was only $750, my bid was actually $2,106 because this is still the only pair of RCA 211 black plates, which I have seen offered for sale.
In my case the difference was night & day, the RCA's made my Shuguangs sound nearly dead with no dynamics. I have found that despite my excellent critical listening ability, I have to do an A / B comparison because my memory of how good or bad a particular is very poor, I remember what I liked, but cannot judge how much better or worse a tube is without a direct comparision.
You were judging input tubes, which might not have as notable a difference as output tubes in an SET circuit.
What are tak tubes? I have not seen or heard of them, can you provide a link? My favorite 300 B's were the Sophie Electric carbon plates, which I actually preferred over the last 300 B's, which Western Electric produced. I had 4 WE's per channel in a push-pull circuit. It was another painful learening experience as I discovered, you can even ruin great tubes by putting them in a circuit, which does not maximize the tubes' performance. I got out nearly whole because the WE's appreciated almost as much as my equipment depreciated. My speakers are only 86db efficient, which means they cannot be adequately driven by a 300 B in an SET circuit.
I am still waiting waiting for a good explanation as to why no one has been able to make new tubes as good as the old NOS tubes.
The best explanation thus far is that the old tubes were made with trace elements, which are either extinct or so rare that they would be cost prohibitive to use in audio vacuum tubes. Please let me know if, you have heard a better explanation.

"In my case the difference was night & day, the RCA's made my Shuguangs sound nearly dead with no dynamics."

Yeah, same with their 6SN7s.

[MylesBAstor]

Ps vane tubes look right but don't sound right. But acceptable in some ways.
I own a pair of the metal base 101 d and a real nos black base with the skinny pins I use with adaptors.
While 80 year old tubes do sound better it's not night and day. Why don't anyone make real new nos is there no demand for them. He picture I posted of the tak tubes seems to be very close to the nos
have you tried any of them.??
Also have you heard any elrogs 300B

Do you have a day of your life to lose? Talk to Tim de Paravicini.

[mep]

A topic I've often pondered. Years ago, we were told the reason tube testers weren't a reliable index was that they needed to be calibrated. Then tubes needed to be tested at normal operating voltages. Then tested in the actual circuit to ensure circuit meets specs. Yet despite this, I find the results from tube testers are disappointing. And certainly no predictor of sound. Or longetivity. (Other than manufacturers and someone like a Kevin Deal, I wonder how many tube retailers do a tempered burn-in?) I won't mention the name but years ago, this measurements oriented person-you know the type, tubes have no sound but it's how they are used in the circuit- was selling these after market upgrade kits for various tube preamps and amps. Tubes measured perfectly but they sounded like expensive crap.

Are tube testers perhaps of more utility when it comes to output compared to small signal tubes? There's no question on a tube tester when you have a bad output tube. Yet small signal tubes can test good yet you know they are shot because they have no life e.g. Dynamics. Just boring sounding. Without question, tube testers have a role in matching tubes, matching triode sections or detecting microphonics. But bad or shot small signal tubes?

Has anyone found a tube tester that really shows when a small signal tube is shot?

Many of the emission type tube testers like Hickok had a cathode life test function where it would drop the heater voltage and measure how much the cathode emission dropped off. The stronger the cathode emission, the less it drops off when the heater voltage is reduced. Unless a tube dies an early death caused by an internal short, bad vacuum (gassy tube), or a complete loss of vacuum (the tube will turn white), the length of a tube's life is predicated on the quality of the cathode. A natural death for a tube is when the cathode is no longer capable of emitting enough electrons to maintain the operating parameters for which the tube was designed to work in.

And has been stated in this thread, different circuits pose different operating conditions for the tubes. Some circuits operate the tubes at their minimum rated voltages and currents while other circuits push tubes right up to their maximum ratings which is a recipe for a short life. An example of an output tube that is currently loafing in most circuits it's used in is the KT-150. An example of an amp that pushed the output tubes to the max limit of plate dissipation was the Quicksilver MS-190. Mike Sanders designed this amp to use no kidding real Mullard EL-34 tubes which exceeded the ratings for "normal" EL-34 tubes. The MS-190 ran the EL-34s right at the maximum plate dissipation rating of 25 watts. I loved my MS-190s, but I simply couldn't find current production EL-34s that could stand up to the punishment. Tube testers don't mimic the circuit the tube will be used in.

[Alrainbow]

Do you have a day of your life to lose? Talk to Tim de Paravicini.

Who is Tim ? I know this is a rabbit hole coming

[MylesBAstor]

Who is Tim ? I know this is a rabbit hole coming

One of the best and most famous designers of tube gear ever. He designed the classic Lux 3045 amp and now does EAR. Plus he's redesigned quite a few recording studios. An encyclopedia of knowledge about tubes. Oh and BTW his redesigned Quad '57s are something to behold. Even from a person who isn't the biggest quad fan.