tardishead
Well-known member
I know this has been touched on before but I wondered if we could make a definitive resource page for this topic.
So the first point is that tubes should be matched within the circuit they are being used for - or if thats not possible in a test circuit which has the same biasing and load parameters.
Tubes should be tested at a few different grid bias voltages - the more the better.
We are talking about dual triodes or pairs of single pentodes(sometimes used as triodes)
In circuits where tubes are paralleled it is not essential for all to match exactly the same but they should be matched in complimentary push pull pairs. For instance 4 dual triodes which are very closely balanced in themselves between each side will work correctly.
Am I right in saying that current draw is the most important parameter to match?
Because if the two sides in a push pull gain reduction circuit have different current draw then this DC unbalance creates thumping in the circuit and poor frequency response in the output transformer. Gain is only worth considering if the two sides draw the same current.
In the case that sides have closely matched current draw but different gains what are the problems associated with this? Intermodulation distortion?
And for that matter - in any type of amplifier circuit?
Obviously gain is important in stereo mastering applications when considered as average gain from the entire gain reduction circuit.
I recently tested a batch of 6BC8 dual triodes.
Out of 20 tubes I really only had one that had a good balance at all operating points.
Most of them did not measure any better than 0.2ma imbalance between sides. That is at most of the operating points, balance was worse than 0.2ma.
And sure enough they did not sound good in the circuit. Trimming the cathode currents between sides did not improve anything.
What tolerance is generally considered workable? Equal or better than 0.05ma? 5%? 1%?
Recently I have read that a lot of people are finding it hard to get balanced 6BC8s in singles let alone batches.
The manufacturer of the Unfairchild limiter has even decided to stop using 6BC8s because of this problem.
Why is this.
Are any brands noted for their balancing of dual triodes?
Is 6BC8 a particular example. Perhaps it was not considered important to balance this particular tube because it was used mostly in a different type of circuit - cascode??
It was designed as a TV tuner tube and this was the primary market drive - any others were considered irrelevant.
I presume that when Datasheets for a particular tube specify use in Automatic Gain Control circuits that it is more likely that the manufacturers put a little more effort into balancing dual triodes.
For instance 6386 and the Russian 6N5P?
OR IS IT JUST THE WAY THE COOKIE CRUMBLES?
Any other points worth considering?
What are your experiences in matching bottles - particularly dual triodes.
So the first point is that tubes should be matched within the circuit they are being used for - or if thats not possible in a test circuit which has the same biasing and load parameters.
Tubes should be tested at a few different grid bias voltages - the more the better.
We are talking about dual triodes or pairs of single pentodes(sometimes used as triodes)
In circuits where tubes are paralleled it is not essential for all to match exactly the same but they should be matched in complimentary push pull pairs. For instance 4 dual triodes which are very closely balanced in themselves between each side will work correctly.
Am I right in saying that current draw is the most important parameter to match?
Because if the two sides in a push pull gain reduction circuit have different current draw then this DC unbalance creates thumping in the circuit and poor frequency response in the output transformer. Gain is only worth considering if the two sides draw the same current.
In the case that sides have closely matched current draw but different gains what are the problems associated with this? Intermodulation distortion?
And for that matter - in any type of amplifier circuit?
Obviously gain is important in stereo mastering applications when considered as average gain from the entire gain reduction circuit.
I recently tested a batch of 6BC8 dual triodes.
Out of 20 tubes I really only had one that had a good balance at all operating points.
Most of them did not measure any better than 0.2ma imbalance between sides. That is at most of the operating points, balance was worse than 0.2ma.
And sure enough they did not sound good in the circuit. Trimming the cathode currents between sides did not improve anything.
What tolerance is generally considered workable? Equal or better than 0.05ma? 5%? 1%?
Recently I have read that a lot of people are finding it hard to get balanced 6BC8s in singles let alone batches.
The manufacturer of the Unfairchild limiter has even decided to stop using 6BC8s because of this problem.
Why is this.
Are any brands noted for their balancing of dual triodes?
Is 6BC8 a particular example. Perhaps it was not considered important to balance this particular tube because it was used mostly in a different type of circuit - cascode??
It was designed as a TV tuner tube and this was the primary market drive - any others were considered irrelevant.
I presume that when Datasheets for a particular tube specify use in Automatic Gain Control circuits that it is more likely that the manufacturers put a little more effort into balancing dual triodes.
For instance 6386 and the Russian 6N5P?
OR IS IT JUST THE WAY THE COOKIE CRUMBLES?
Any other points worth considering?
What are your experiences in matching bottles - particularly dual triodes.