I don't deny the coolness factor, but there are sound, rational reasons to use tubes in many places in a signal chain. Why would you make up something like that?
Hi,
Hmm, as you emphasised in prior posts of this thread how chokes cause non-linearity and such, could you suggest the exact way in which any of your tube circuits are superior in engineering terms at conventional signal levels, compared to - say an OPA604 Op-Amp with external class A current booster, run on +/-24V rails?
Ciao T
Hmm, as you emphasised in prior posts of this thread how chokes cause non-linearity and such, could you suggest the exact way in which any of your tube circuits are superior in engineering terms at conventional signal levels, compared to - say an OPA604 Op-Amp with external class A current booster, run on +/-24V rails?
Ciao T
I was under the impression that "coolness" summed up your attraction to tubes. Please forgive me if i have interpreted this wrong.
The attraction of vacuum tubes / valves.
Last edited:
I never said "nonlinearity." You're making that up.
To do what? Do you usually design circuits in isolation, without consideration of system issues?
You're forgiven. Like I said, coolness is not to be denied, but there are engineering reasons as well.
Hello Thorsten. You wrote:
"...Either had ridiculous levels of inductance and very low interwinding capacitance (I posted the S&B values on line but cannot now find them). I never tested Magnequest, but have to assume that they too work well enough with low capacitance and high inductance."
yes and yes as re: MQ grid chokes.
L will vary with chosen core material---- all of our published L's are stated conservatively.
winding capacitance is approx 25 puffs on our first gen grid chokes and approx 14 puffs
on our second (current) generation of grid chokes.
from the perspective of capacitance... the second gen grid chokes present less capacitance
to the driving source than the input capacitance of the newly developed KT-120 tube.
MSL
"...Either had ridiculous levels of inductance and very low interwinding capacitance (I posted the S&B values on line but cannot now find them). I never tested Magnequest, but have to assume that they too work well enough with low capacitance and high inductance."
yes and yes as re: MQ grid chokes.
L will vary with chosen core material---- all of our published L's are stated conservatively.
winding capacitance is approx 25 puffs on our first gen grid chokes and approx 14 puffs
on our second (current) generation of grid chokes.
from the perspective of capacitance... the second gen grid chokes present less capacitance
to the driving source than the input capacitance of the newly developed KT-120 tube.
MSL
Last edited:
In some applications like an SE amplifier using a 50 (dht) the only viable choices are a grid choke or IT.. Not even a D3A will drive a 10K load to the required levels. (grid resistor 10K max) Some people will tell you that a 50 under the right circumstances sounds better than some big name 300B in the same application.
This is just one instance where a grid choke or IT would be superior to a resistor for practical engineering based reasons - reducing onerous drive requirements to something manageable in the case of a 50, ever try to deliver 18mApp into a 10K load with any readily available tube out there?
FWIW I am enjoying the conversation, please keep it up!
Hi Sy. You wrote in part;
"....ex cathedra pronouncements by those with something to sell or promote,..."
This is an unfortunate statement... it's a red herring. The arguments made should stand or fall independently of the person making them... whether they do or don't have a commercial interest in the products being discussed.
Often times the best applications assistance can be gained by consulting with the company or person who actually has designed and manufactured a given component. That manufacturer may be moreso aware of it's weaknesses and strengths.
Sy speaking;
"... as well as a rational reason why plate chokes were useful in limited situations, but not in most others."
I must have missed the post where this conclusion was reached and stated.
Sy speaking;
"I well understand that this disagrees with the fashion statements that many like to promote..."
I surely hope that your not pointing at ALL manufacturers. I have been on record and have posted many times across many different forums a sentiment
that contrasts with that overly broad brush stroke....
once again... I'll state... in my opinion.... most consumers would be better off by focusing on obtaining quality magnetic components in the KEY circuit positions first.
My recommendation or suggestion holds especially true if the builder has a limited budget to spend on iron. I'd rather see them, for example, buy a great quality output transformer and forego use of an input, TVC, interstage, plate choke, grid choke, etc... so that they could get a better quality output trans... rather than diluting their magnetic dollars by stretching those dollars over a half a dozen components....
If your using a tube output stage (at the power amp level) your most likely going to need an output transformer... buying a fifty dollar output so that you can also buy some $50 interstages and some $50 plate chokes and some $50 grid chokes....
doesn't (in this case) make (again, to me) as much sense as taking all those fifty dollars and bundling them up together to buy the best outputs that those aggregate dollars could get you.
This is not the viewpoint of someone trying to jam grid chokes or plate chokes down the consumers' throat.
MSL
Last edited:
Mike, FWIW, I wasn't talking about you. Although you sell chokes, I haven't seen you make (at least here) unsupported performance claims. As for plate chokes, I have yet to see any evidence that, high swing applications aside, they work any better than smaller, lighter, cheaper CCS loads. Or even as well.
I can see the IT (though IMO it's usually not a good choice, YMMV), but not the grid choke. How can a grid choke reduce drive current requirements? Presumably, its impedance is some orders of magnitude higher than the grid impedance when the grid is drawing current. And unlike a plate choke, the grid choke isn't carrying current, so it won't provide a boost in swing.
Let's see, 18 mA p-p into 10k is 180Vpp or 63VRMS. Yeah, you'd want either an IT or (my choice) a MOSFET. Though a 6528 could work...
I can see the IT (though IMO it's usually not a good choice, YMMV), but not the grid choke. How can a grid choke reduce drive current requirements? Presumably, its impedance is some orders of magnitude higher than the grid impedance when the grid is drawing current. And unlike a plate choke, the grid choke isn't carrying current, so it won't provide a boost in swing.
Let's see, 18 mA p-p into 10k is 180Vpp or 63VRMS. Yeah, you'd want either an IT or (my choice) a MOSFET. Though a 6528 could work...
I can't say I've ever seen any reputable transformer manufacturer make such claims. Indeed, when I tried to order some plate chokes from Jack at Electra-Print, he spent the better part of the phone call trying to convince me that a resistor works better.
This is surely true. As you say, the reason for using a plate choke is because it can swing a lot of volts. It is a trade off where one does not want to pay for the overhead of a much higher B+ and CCS that can operate at such volts. The choke is a good approximation for less money and less danger.
Hi Sy,
I would have expected someone of your erudition to at least check the facts before posting, if not actually knowing beforehand. Alas, I have been proven wrong.
For your reference, the type 50 Triode from RCA is listed with a maximum grid circuit DC resistance of 10K.
A 4,000H Choke with a 12pF stray capacitance and 1000 Ohm DCR actually provides a grid circuit DC Resistance of much less than the stated limit while offering an AC Impedance of over 250KOhm from 10Hz to 50KHz. This AC Impedance demands a MUCH lower AC shunt current than a 10K gridresistor.
In fact, the AC Impedance is still appreciably higher than the 100K DC Resistance limit given for the grid circuit of the 300B in fixed bias mode, so even here the AC current demand from the driver stage is reduced.
So, applyig some actual research and doing your sums is recommended before posting.
Ciao T
I would have expected someone of your erudition to at least check the facts before posting, if not actually knowing beforehand. Alas, I have been proven wrong.
For your reference, the type 50 Triode from RCA is listed with a maximum grid circuit DC resistance of 10K.
A 4,000H Choke with a 12pF stray capacitance and 1000 Ohm DCR actually provides a grid circuit DC Resistance of much less than the stated limit while offering an AC Impedance of over 250KOhm from 10Hz to 50KHz. This AC Impedance demands a MUCH lower AC shunt current than a 10K gridresistor.
In fact, the AC Impedance is still appreciably higher than the 100K DC Resistance limit given for the grid circuit of the 300B in fixed bias mode, so even here the AC current demand from the driver stage is reduced.
So, applyig some actual research and doing your sums is recommended before posting.
Ciao T
Hi Sy,
I do need to take you task. In a post above I requested your engineering reasons.I will repeat my request:
I will in fact be more precise in my request.
Please provide the reasoning for the ENGINEERING DRIVEN decision for using:
a) A transformer, with its limitations and very high cost instead of a suitable much higher performance and much lower cost solid state circuit for balanced to single ended conversion
b) A tube, with it's high distortion and high cost, instead of using a suitable much higher performance and lower solid state circuit for the follower
in your "heretical line stage".
Enquiring minds want to know.
You know, for me, a "because I like them better" would be good enough. But I would never claim that as decision that was engineering driven, probably rather the opposite.
Please also note that I would find your heretical linestage very uninteresting had it been designed based on engineering driven logic, with (say) a INA134 as balanced to SE converter and a OPA604 as follower.
Even though I can assure in measured terms this little solid state contraption would wipe the floor with the "heretical linestage" as you designed it at a fraction of the cost.
Ciao T
I do need to take you task. In a post above I requested your engineering reasons.I will repeat my request:
I will in fact be more precise in my request.
Please provide the reasoning for the ENGINEERING DRIVEN decision for using:
a) A transformer, with its limitations and very high cost instead of a suitable much higher performance and much lower cost solid state circuit for balanced to single ended conversion
b) A tube, with it's high distortion and high cost, instead of using a suitable much higher performance and lower solid state circuit for the follower
in your "heretical line stage".
Enquiring minds want to know.
You know, for me, a "because I like them better" would be good enough. But I would never claim that as decision that was engineering driven, probably rather the opposite.
Please also note that I would find your heretical linestage very uninteresting had it been designed based on engineering driven logic, with (say) a INA134 as balanced to SE converter and a OPA604 as follower.
Even though I can assure in measured terms this little solid state contraption would wipe the floor with the "heretical linestage" as you designed it at a fraction of the cost.
Ciao T
Guess you didn't read that part.
In any case, if you can't design a stage to drive a 10k impedance so that you can use an oddball pathological tube with (let's face it) mediocre linearity then you ought not to be designing tube circuits. It also begs the question, what's the reason for the 10k limitation? Grid current? Nonlinear with signal? Well, doesn't your driver still have to cope with that, grid choke or no? Wouldn't the same design criteria that reduce distortion at that interface make the stage more suitable for dealing with a 5 cent resistor having an order or two of magnitude better stray capacitance?
They key is to think of the system design, otherwise you end up doing pathological circuits to accord with the dictates of fashion or marketing.
Hi Sy,
I did. I guess you did not the part of your post that reads:
In order to enlighten you non the point that clearly escaped you, I pointed out that in the specific cited case a gridchoke can reduce the drive current in fact by a factor 25 or more over the audio range.
Just because I CAN design such a stage easily does not mean I actually WANT TO USE such a stage.
It also begs the question, what's the reason for the 10k limitation? [/QUOTE]
Farly poor vacuum in this early tube caused significant grid emission even at quiescent conditions.
Surely someone who likes to teach others about Engineering Driven resoning does not need to ASK such questions? Surely you can research them if you want to know the answers? Surely you do not need a snakeoil merchand to give you the engineering driven reasons?
Absolutely. Funny that I see such very much in your circuits, while you do in mine... Except my circuit tricks where fashion 50 or more years ago and are long out of fashion, while yours (CCS, Mosfets et al) are very recent fashions and fads, if you allow me the observation.
In the end we both look at things from a system design level, microphone to speaker, so that is good and we agree on this, even if we disagree on how to actually do this.
Ciao T
PS, I am still waiting for you revealing your low distortion loudspeakers and your engineering reasons for transformers and tubes...
I did. I guess you did not the part of your post that reads:
In order to enlighten you non the point that clearly escaped you, I pointed out that in the specific cited case a gridchoke can reduce the drive current in fact by a factor 25 or more over the audio range.
Just because I CAN design such a stage easily does not mean I actually WANT TO USE such a stage.
It also begs the question, what's the reason for the 10k limitation? [/QUOTE]
Farly poor vacuum in this early tube caused significant grid emission even at quiescent conditions.
Surely someone who likes to teach others about Engineering Driven resoning does not need to ASK such questions? Surely you can research them if you want to know the answers? Surely you do not need a snakeoil merchand to give you the engineering driven reasons?
Absolutely. Funny that I see such very much in your circuits, while you do in mine... Except my circuit tricks where fashion 50 or more years ago and are long out of fashion, while yours (CCS, Mosfets et al) are very recent fashions and fads, if you allow me the observation.
In the end we both look at things from a system design level, microphone to speaker, so that is good and we agree on this, even if we disagree on how to actually do this.
Ciao T
PS, I am still waiting for you revealing your low distortion loudspeakers and your engineering reasons for transformers and tubes...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.