http://www.maximatcher.com/
I recently got myself a reliable Stark 9-66 so i wouldent even consider blowing 5 or 6 bones on one of these but has anyone tried or heard of these units?
I recently got a set of EL84's with different systems and the seller who supposedly has a good reputation in another online community said his maxi tester tests the tubes ability to amplify signal and the quad he sent me matched up.
meanwhile on my stark there are two (with disc style getters.. old rft's) that test noticable lower then the halo getter rft's he sent.
what would be more reliable.. a solid, recently calibrated stark or hickok mutual conductance tester or one of these new fangled digital units?
I recently got myself a reliable Stark 9-66 so i wouldent even consider blowing 5 or 6 bones on one of these but has anyone tried or heard of these units?
I recently got a set of EL84's with different systems and the seller who supposedly has a good reputation in another online community said his maxi tester tests the tubes ability to amplify signal and the quad he sent me matched up.
meanwhile on my stark there are two (with disc style getters.. old rft's) that test noticable lower then the halo getter rft's he sent.
what would be more reliable.. a solid, recently calibrated stark or hickok mutual conductance tester or one of these new fangled digital units?
Hi smak,
Your Stark 9-66 is a Hickok Dynamic Mutual Conductance tester. It's made under license from Hickok. Says so right on the front panel.
Those two above will test far more tubes than the new one you indicated. The digital scale is sexy, but misleading maybe. The characteristics will vary with current and voltage. So use your Stark and be happy.
If you want to be more accurate, build the circuit and measure the prospective tubes in that circuit. That's more accurate.
Regarding the two tubes. If the Stark says they are different, they are different in transconductance.
-Chris
Your Stark 9-66 is a Hickok Dynamic Mutual Conductance tester. It's made under license from Hickok. Says so right on the front panel.
Those two above will test far more tubes than the new one you indicated. The digital scale is sexy, but misleading maybe. The characteristics will vary with current and voltage. So use your Stark and be happy.
If you want to be more accurate, build the circuit and measure the prospective tubes in that circuit. That's more accurate.
Regarding the two tubes. If the Stark says they are different, they are different in transconductance.
-Chris
yeah i know its mutual conductance.. i guess i could have used another comma there.
just was curious about the discrepency between what the seller says his new aged tester reads vs what my mutual conductance tester shows.
is transconductance in direct relation to the amplification of a vacuum tube?
not sure how its possible for a tubes to read weaker/stronger on my stark and the same on a "maxi"
just was curious about the discrepency between what the seller says his new aged tester reads vs what my mutual conductance tester shows.
is transconductance in direct relation to the amplification of a vacuum tube?
not sure how its possible for a tubes to read weaker/stronger on my stark and the same on a "maxi"
It's all cobblers...
It's easy for two perfectly good valve testers to give different results on the same valves. This business of "matched" valves is pretty much a fiction. Matched for what? Mutual conductance? Amplification factor? Anode current? And under what conditions? Remember that mutual conductance and amplification factor are small-signal parameters that change with operating conditions. My AVO VCM163 measures mutual conductance by injecting a small 15kHz signal into the grid and seeing what comes out at the anode, so it can measure mutual conductance under any operating condition it can set. Thus, I could "match" valves for mutual conductance at one set of operating conditions and find that they're different under another.
Of course, you could always use a curve tracer and look for pairs of valves with perfectly overlaying traces. Out of one hundred valves, you might well find a pair. But what would be the point? Why do you need matched valves? If you had a push-pull output stage that didn't have a mean of balancing or auto-balancing anode currents in the output transformer, you'd quite like the valves to be matched for anode current. But how do you know that the valves will age in the same way? You don't.
So why do so many vendors sell "matched" valves? The fact of the matter is that manufacturing valves is high technology, and even when the manufacturers were making enough to know what they were doing, there was still quite a spread. That didn't matter because the designers of the time knew what they were dealing with, and by and large, they designed their circuits to cope. These days, insufficient valves are made for the manufacturers to gain good experience and many are made in countries where their experience of quality control is, cough, cough, somewhat limited. Further, the cheap availability of DVMs etc has unreasonably raised the expectations of users. So what do the vendors of current production do? They run all the valves through a massive computerised measurement system to weed out the out-and-out rejects and then they have the data to be able to "match" valves and charge you a premium for doing so.
Worse, a small vendor may pop a few valves onto a simple tester, dial in the conditions it says in the tester's code book and pronounce valves to be "matched" because two valves drive a meter needle or DVM to the same point. They probably have no knowledge of what they are doing. Someone who really knew what they were doing would ask you how the valves were going to be used and would then warn that their "matching" couldn't be all that good.
You'd do far better checking that when your EL84 are in circuit that each valve in a pair draws just about the same anode current than fooling around with a valve tester.
Sorry for the rant, but valve testers are, and always have been, dodgy bits of equipment that need very cautious interpretation of their results.
It's easy for two perfectly good valve testers to give different results on the same valves. This business of "matched" valves is pretty much a fiction. Matched for what? Mutual conductance? Amplification factor? Anode current? And under what conditions? Remember that mutual conductance and amplification factor are small-signal parameters that change with operating conditions. My AVO VCM163 measures mutual conductance by injecting a small 15kHz signal into the grid and seeing what comes out at the anode, so it can measure mutual conductance under any operating condition it can set. Thus, I could "match" valves for mutual conductance at one set of operating conditions and find that they're different under another.
Of course, you could always use a curve tracer and look for pairs of valves with perfectly overlaying traces. Out of one hundred valves, you might well find a pair. But what would be the point? Why do you need matched valves? If you had a push-pull output stage that didn't have a mean of balancing or auto-balancing anode currents in the output transformer, you'd quite like the valves to be matched for anode current. But how do you know that the valves will age in the same way? You don't.
So why do so many vendors sell "matched" valves? The fact of the matter is that manufacturing valves is high technology, and even when the manufacturers were making enough to know what they were doing, there was still quite a spread. That didn't matter because the designers of the time knew what they were dealing with, and by and large, they designed their circuits to cope. These days, insufficient valves are made for the manufacturers to gain good experience and many are made in countries where their experience of quality control is, cough, cough, somewhat limited. Further, the cheap availability of DVMs etc has unreasonably raised the expectations of users. So what do the vendors of current production do? They run all the valves through a massive computerised measurement system to weed out the out-and-out rejects and then they have the data to be able to "match" valves and charge you a premium for doing so.
Worse, a small vendor may pop a few valves onto a simple tester, dial in the conditions it says in the tester's code book and pronounce valves to be "matched" because two valves drive a meter needle or DVM to the same point. They probably have no knowledge of what they are doing. Someone who really knew what they were doing would ask you how the valves were going to be used and would then warn that their "matching" couldn't be all that good.
You'd do far better checking that when your EL84 are in circuit that each valve in a pair draws just about the same anode current than fooling around with a valve tester.
Sorry for the rant, but valve testers are, and always have been, dodgy bits of equipment that need very cautious interpretation of their results.
I am agreement with EC8010...
A few things to keep in mind about matching...
Most places that match tubes are only using DC current to match tubes...which does not nessecerily mean the gm is matched...
While it is imortant to have the DC current matched, it is of most importance to have the gm matched.. ANother thing to keep in mind..many DC current matching stuff is actually looking at the cathode current, not plate current...this can be a problem since some modern production tubes can have excessive screen currents...
The problem with most valve testers, is that the voltages are not well regulated... ALL voltages need to be rock solid for acurate gm measurements...One problem I noticed at some testing places is that the heaters were not regulated...so when you have 2 tubes in line for testing the heater voltage will vary according to how many tubes are lined up in these big testers...such as having 20 tubes in line.... I have seen the heaters vary from 6.1 to 6.9v depending on the number of tubes in the tester...
I have seen bias voltages change under tester loading...
Tube testing is quite simple and repeatable, provided you follow the rules... ALL voltage must be well regulated including the heaters... The parameters such as mu, gm..ect are all derived from partial differential equations, which shows that ALL other constants must maintain constant...
Now comes to the important part...lets say you have "matched" valves....matched to what??? They could be matched butthe gm could be in the gutter.... Are forefathers set the standards that we should adhere to.... If you look in the tube manual, the specs are right there for each valve... They will specify the conditions of the test, such as all the voltages and then it will specify the gm at that test point... For example pentodes, tetrodes are mostly tested at the standard 250V plate, 250V screen....
Now keep in mind some argue that the valves should be measured at the voltages it will see in the amp such as 450V to 500V range...
These voltages are typical in the P-P arrangement, while the valves may IDLE at these higher voltages....the dynamic operation will swing the plate from idle voltage to close to 0 voltage or just at the knee... 250V is roughly the mid point in the dynamic AC voltage swing and the small signal gm at this point is the average of the waveform which is fairly close to the large signal gm for that operation.....
If everyone universally used the same standards for testing valves, it would be a heck of a lot easier...
The tube tester I build use solid state voltage regulators.. The plate voltage is adjustable with a single knob from 0 to 600V with digital readout...same with the screen Voltage.... The bias is regulated with digital readout as well... The AC signal to the grid to adjustable and read out in digital meter... The plate has a digitial ammeter in series with it to read the gm with respect to the grid drive voltage... The screen has DC ammeter to check screen current as well..
I also have a adjustable heater voltage with digitial readout...
I can adjust to any condition for testing, with confidence in the measurements.... i simply set the voltages to the tube manual specs..
A few things to keep in mind about matching...
Most places that match tubes are only using DC current to match tubes...which does not nessecerily mean the gm is matched...
While it is imortant to have the DC current matched, it is of most importance to have the gm matched.. ANother thing to keep in mind..many DC current matching stuff is actually looking at the cathode current, not plate current...this can be a problem since some modern production tubes can have excessive screen currents...
The problem with most valve testers, is that the voltages are not well regulated... ALL voltages need to be rock solid for acurate gm measurements...One problem I noticed at some testing places is that the heaters were not regulated...so when you have 2 tubes in line for testing the heater voltage will vary according to how many tubes are lined up in these big testers...such as having 20 tubes in line.... I have seen the heaters vary from 6.1 to 6.9v depending on the number of tubes in the tester...
I have seen bias voltages change under tester loading...
Tube testing is quite simple and repeatable, provided you follow the rules... ALL voltage must be well regulated including the heaters... The parameters such as mu, gm..ect are all derived from partial differential equations, which shows that ALL other constants must maintain constant...
Now comes to the important part...lets say you have "matched" valves....matched to what??? They could be matched butthe gm could be in the gutter.... Are forefathers set the standards that we should adhere to.... If you look in the tube manual, the specs are right there for each valve... They will specify the conditions of the test, such as all the voltages and then it will specify the gm at that test point... For example pentodes, tetrodes are mostly tested at the standard 250V plate, 250V screen....
Now keep in mind some argue that the valves should be measured at the voltages it will see in the amp such as 450V to 500V range...
These voltages are typical in the P-P arrangement, while the valves may IDLE at these higher voltages....the dynamic operation will swing the plate from idle voltage to close to 0 voltage or just at the knee... 250V is roughly the mid point in the dynamic AC voltage swing and the small signal gm at this point is the average of the waveform which is fairly close to the large signal gm for that operation.....
If everyone universally used the same standards for testing valves, it would be a heck of a lot easier...
The tube tester I build use solid state voltage regulators.. The plate voltage is adjustable with a single knob from 0 to 600V with digital readout...same with the screen Voltage.... The bias is regulated with digital readout as well... The AC signal to the grid to adjustable and read out in digital meter... The plate has a digitial ammeter in series with it to read the gm with respect to the grid drive voltage... The screen has DC ammeter to check screen current as well..
I also have a adjustable heater voltage with digitial readout...
I can adjust to any condition for testing, with confidence in the measurements.... i simply set the voltages to the tube manual specs..
Geek's method of matching tubes with DC grid-shift....
- Apply CCS to the plate appropriate for the tube.
- Measure plate + screen current (analog meter is in series with the CCS) at zero bias.
- Moving the grid negative in 5 volt increments, shift the grid and note the current reading.
- Continue until grid-shift results in a plate current of less than 1mA (or 100uA if 1 or 3mA Ia was used, as in the case of 12A_7 series).
- If the readings differed throught by less than 3%, declare them statically matched.
- Apply CCS to the plate appropriate for the tube.
- Measure plate + screen current (analog meter is in series with the CCS) at zero bias.
- Moving the grid negative in 5 volt increments, shift the grid and note the current reading.
- Continue until grid-shift results in a plate current of less than 1mA (or 100uA if 1 or 3mA Ia was used, as in the case of 12A_7 series).
- If the readings differed throught by less than 3%, declare them statically matched.
wow you guys really know how to go off on a tangent eh? ;P i guess thats a no about encountering the maxi tester.
some good info though.
by "matched" i'd expect people want tubes that have been measured on a mutual conductance tester and the readings are within 10% of each other.
of course you can match to anything relative.. colour of base, emissions tester readings, style of internal components etc.. but when a good seller advertises "matched" tubes i'd assume its for use in machinery that doesent allow to adjust any bias for tubes out of tune to each other.
for me, yeah.. its a push-pull ultralinear hifi amp with no bias adjust. so i'm guessing anode current is the ideal characteristic i'd like to match but a mutual conductance reading is better than nothing right?
is it possible that this maxi tester uses anode current to match tubes and this accounts for the difference between unmatching tubes in my tester and matching in the maxi?
just to clarify, the el84's i got are two styles of RFT's.. 2 are regular halo getters but 2 have flat disc getters and measure a bit less than the halo getters in transconductance.
some good info though.
by "matched" i'd expect people want tubes that have been measured on a mutual conductance tester and the readings are within 10% of each other.
of course you can match to anything relative.. colour of base, emissions tester readings, style of internal components etc.. but when a good seller advertises "matched" tubes i'd assume its for use in machinery that doesent allow to adjust any bias for tubes out of tune to each other.
for me, yeah.. its a push-pull ultralinear hifi amp with no bias adjust. so i'm guessing anode current is the ideal characteristic i'd like to match but a mutual conductance reading is better than nothing right?
is it possible that this maxi tester uses anode current to match tubes and this accounts for the difference between unmatching tubes in my tester and matching in the maxi?
just to clarify, the el84's i got are two styles of RFT's.. 2 are regular halo getters but 2 have flat disc getters and measure a bit less than the halo getters in transconductance.
Hi smak,
Just use the pairs as pairs.
Tube testers were designed to help a service tech find a defective tube. They were never designed to provide matching for tubes. As EC8010 pointed out, that is the job of a curve tracer. The easiest way for us poor mortals to check balance is to stick the tubes in a circuit and measure it.
The match a good tester indicates are for those exact conditions. Milage in your circuit will probably vary.
Tubes deviate from their "bogey" values, but much less than transistors do.
-Chris
Just use the pairs as pairs.
Tube testers were designed to help a service tech find a defective tube. They were never designed to provide matching for tubes. As EC8010 pointed out, that is the job of a curve tracer. The easiest way for us poor mortals to check balance is to stick the tubes in a circuit and measure it.
The match a good tester indicates are for those exact conditions. Milage in your circuit will probably vary.
Tubes deviate from their "bogey" values, but much less than transistors do.
-Chris
Hi smak,
I don't believe the maxi tester is of value to you over what the Stark / AVO / Hickok will provide.
Some experimenting by yourself would be of much higher value to you. Understanding how tubes react will allow you to relax a little. The real world seldom performs as you would wish, relax and accept this.
-Chris
It's terribly expensive. There are members working on projects that do this.
No, the results are displayed on a screen, or data is taken and stored in an array in a computer, then processed to display on the screen or printer.
I don't believe the maxi tester is of value to you over what the Stark / AVO / Hickok will provide.
Some experimenting by yourself would be of much higher value to you. Understanding how tubes react will allow you to relax a little. The real world seldom performs as you would wish, relax and accept this.
-Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.