andyjevans,
Yes, I was using the wrong term. I was unaware of the term "Filament Bias"
Plate current:
As to biasing the tube plate current: I have used fixed grid bias (but adjustable to get the desired current), self bias, and Battery bias.
All have tradeoffs.
I have used AC, DC regulators, and DC with brute force CRC to heat 300B filaments.
The CRC had 1 to 2 mV of ripple, which translates to hum not being an issue for me.
For a theoretical worst case, if one end of the tube filament was grounded, and all of the filament cloud of electrons were at the other end:
(2 mV * 3.8) / 25 = 308 uV hum. That is OK for my speakers.
But the cloud of electrons of the tube filament is not all at one end, the gain is less than u, and the 5000:8 transformer divides by 25.
I get less than 100uV of hum.
Yes, I was using the wrong term. I was unaware of the term "Filament Bias"
Plate current:
As to biasing the tube plate current: I have used fixed grid bias (but adjustable to get the desired current), self bias, and Battery bias.
All have tradeoffs.
I have used AC, DC regulators, and DC with brute force CRC to heat 300B filaments.
The CRC had 1 to 2 mV of ripple, which translates to hum not being an issue for me.
For a theoretical worst case, if one end of the tube filament was grounded, and all of the filament cloud of electrons were at the other end:
(2 mV * 3.8) / 25 = 308 uV hum. That is OK for my speakers.
But the cloud of electrons of the tube filament is not all at one end, the gain is less than u, and the 5000:8 transformer divides by 25.
I get less than 100uV of hum.
Yes - filament bias is a bit off the beaten path. Not usually in any of the textbooks.
How did you find the choice between
1. Fixed bias?
2. Self bias with a cathode bypass cap, AC or DC?
I'll have to make this real-world choice if I'm ever going to use 300b outputs. I've used AC + bypass cap in the past but not mad about that choice.
Member
Joined 2009
Paid Member
I've been interested in having an all-DHT amp before, with 2A3 at the output but I can't see how to achieve that without going to a 3 stage amp (I'm not wanting to use an IT and would prefer dc coupled) as there aren't any DHTs with enough gain. Of course, this is not based on experience but all the 'bad press' about 3 stage amps and that 2 stage is much better. I have a single JJ-2A3 and Sowter OPT gathering dust, designated for a dc-coupled mono amp. Driver choices with direct coupled IDHT for a 2-stage amp are plentiful. How would you go about putting together an all-DHT amp with a 2A3 ? It gets easier if we go hybrid, using a folded cascode off the driver tube with a PNP transistor of course as we can 'borrow' the gm of the bjt.
Way before that, I think around 1998 or 1999 first shown in a Sound Practice magazine.
B R
Thomas
Yes, depending on how I tweak the physical circuit I will sometimes get better vocals and less bass or more bass and less vocals.
Different coupling caps in particular seem to play a minor, yet considerable, effect.
Like I said earlier, when I get home I will measure some voltages and draw up a schematic.
------------
and to be fair to the others, I am finding what they are saying very interesting. Its pretty easy to learn how to build an amp, but learning how different types of circuits effect the sound is a little harder to research. So its all good in my book.
Plus I am sure that when I do post a schematic, someone will probably find a glaring issue and tell me to go fix it. : P
Last edited:
I just wanna say, I am working on drawing this schematic, and I have run into some things that maybe be causing a problem.
I am going to talk to my friend and see if I can't sort some of this out before I post a final schematic.
I am going to talk to my friend and see if I can't sort some of this out before I post a final schematic.
I´m happy with ECC803 under gyrator. Gives hi ~85x gain, low out Z, and all that with unbypassed Rk. Sounds detailed, nice "universal gain block". Wima polyprops put everywhere, fixed sound issues.
Hi,
What exactly are your speakers? Have you tried it in SE pentode with local feedback? Maybe your friend could mod it for you f you're not really happy..
Maybe use a 600H (preferably nickel core) plate choke (or ccs or gyrator of choice) on top of an 841 dht, filament biased, DC coupled to the 2A3. "Full monkey" or any modern take on L-W.. Then a separate preamp like Ale's 01a for those amps.
Last edited:
x hpter
The charateristic are quite the same, look on JJ site .
The solution is interesting because you can use a 2.5 volt on filaments with less trouble on noise; the driving is little bit better than standard 300B
I have used them many times and are very fine.
http://www.multitask.it/300b/20131217_181644.jpg
the monos are with 2A3-40.
In other amps I used the 300B with only a bias adjustment
Walter
The charateristic are quite the same, look on JJ site .
The solution is interesting because you can use a 2.5 volt on filaments with less trouble on noise; the driving is little bit better than standard 300B
I have used them many times and are very fine.
http://www.multitask.it/300b/20131217_181644.jpg
the monos are with 2A3-40.
In other amps I used the 300B with only a bias adjustment
Walter
In my working point 2A3 should have 64,5V bias; 300B 69V
I measured 62,5V. 2A3JJ must have higher mu
(314V, ~36mA)
Longer distance A-G (just a little here) will increase µ.
I´m afraid, the person who made DS, just copy-pasted graphs/data..my explanation for confusion.
I measured 62,5V. 2A3JJ must have higher mu
(314V, ~36mA)
Longer distance A-G (just a little here) will increase µ.
I´m afraid, the person who made DS, just copy-pasted graphs/data..my explanation for confusion.
Last edited:
What are the options for reliable and sounds good as far as 2A3 current production goes? There are far more options with 300B when it comes to those two criteria. I'm not convinced 300B has any negative sound people associate with it, when I read those it seems more like its insufficiently driven than any fault of the tube.
andyjevans,
Here are my observations on different bias options. I hope these are helpful to some others too.
I will discuss SE amps I built below (PP amps I built would be another discussion).
Self Bias:
45, 2A3, 300B
These all sounded good.
The stereo 45s I used at work, and got lots of good comments, but they were only played softly.
The 2A3s and 300Bs were played at home, and liked by those who stopped by to listen.
Hint: for good bass response, use large capacitance bypass caps across the self bias resistor.
Hint: You should use a self bias resistor and a bypass cap for each tube (more parts).
Advantages:
This is the easiest to implement and get to a desired current range.
There is no thermal runaway, and they can use higher resistance control grid resistors (lighter load on the driver tube).
The left and right channels have nearly the same current, unless the tubes are very badly matched.
As the line voltage varies, the plate current remains fairly constant (versus battery bias).
Disadvantages:
The output power is reduced versus fixed bias and battery bias, unless the B+ is raised by the amount of the bias voltage.
Fixed Bias:
EL34 Triode Wired
These sounded good, and had good reports at various venues.
Unfortunately, I did not use fixed bias (from a power mains supply) on DHTs.
(Battery Bias is also fixed, but not from power mains; See battery bias below).
I will have to try fixed bias on DHTs in amps that are self bias, but which could use more B+.
Advantages:
All the B+ is used for plate to filament, or plate to cathode voltage (not for self bias voltage).
As the power line voltage increases, so does the B+, but the bias also increases, resulting in a fairly constant plate current.
Disadvantage:
You do have to individually adjust the bias on each tube.
At power up, the bias voltage takes time to rise (if there is lots of hum filtering). This works for Indirect Heated Tubes, but not as well for DHTs if the B+ rises quickly. It can result in large inrush current for DHTs, unless a B+ rectifier with a cathode is used.
Battery Bias:
2A3, 300B
I used 9V batteries in series. The driver coupling cap connected to the battery +, the battery - to the grid stopper resistor, the grid stopper resistor to the grid. The junction of the coupling cap to the battery + also connected to the grid return resistor to ground. This way, there is no current in the battery, unless the driver signal is causing grid current (not operated this way, it is in clipping when grid current is drawn).
Advantages:
More Power with a fixed B+
No large inrush current of the output tube at power turn-on.
No large inrush current on a hot start.
Disadvantages:
Batteries should be replaced after 2 or 3 years.
Batteries take space and are heavy.
When the power line voltage increases, B+ increases, so with the constant battery bias, the plate current goes up.
Battery bias voltage is in large increments for medium to high transconductance tubes.
1.5V bias change on a 5,500 micro-mho tube causes 8.25mA plate current change.
Here are my observations on different bias options. I hope these are helpful to some others too.
I will discuss SE amps I built below (PP amps I built would be another discussion).
Self Bias:
45, 2A3, 300B
These all sounded good.
The stereo 45s I used at work, and got lots of good comments, but they were only played softly.
The 2A3s and 300Bs were played at home, and liked by those who stopped by to listen.
Hint: for good bass response, use large capacitance bypass caps across the self bias resistor.
Hint: You should use a self bias resistor and a bypass cap for each tube (more parts).
Advantages:
This is the easiest to implement and get to a desired current range.
There is no thermal runaway, and they can use higher resistance control grid resistors (lighter load on the driver tube).
The left and right channels have nearly the same current, unless the tubes are very badly matched.
As the line voltage varies, the plate current remains fairly constant (versus battery bias).
Disadvantages:
The output power is reduced versus fixed bias and battery bias, unless the B+ is raised by the amount of the bias voltage.
Fixed Bias:
EL34 Triode Wired
These sounded good, and had good reports at various venues.
Unfortunately, I did not use fixed bias (from a power mains supply) on DHTs.
(Battery Bias is also fixed, but not from power mains; See battery bias below).
I will have to try fixed bias on DHTs in amps that are self bias, but which could use more B+.
Advantages:
All the B+ is used for plate to filament, or plate to cathode voltage (not for self bias voltage).
As the power line voltage increases, so does the B+, but the bias also increases, resulting in a fairly constant plate current.
Disadvantage:
You do have to individually adjust the bias on each tube.
At power up, the bias voltage takes time to rise (if there is lots of hum filtering). This works for Indirect Heated Tubes, but not as well for DHTs if the B+ rises quickly. It can result in large inrush current for DHTs, unless a B+ rectifier with a cathode is used.
Battery Bias:
2A3, 300B
I used 9V batteries in series. The driver coupling cap connected to the battery +, the battery - to the grid stopper resistor, the grid stopper resistor to the grid. The junction of the coupling cap to the battery + also connected to the grid return resistor to ground. This way, there is no current in the battery, unless the driver signal is causing grid current (not operated this way, it is in clipping when grid current is drawn).
Advantages:
More Power with a fixed B+
No large inrush current of the output tube at power turn-on.
No large inrush current on a hot start.
Disadvantages:
Batteries should be replaced after 2 or 3 years.
Batteries take space and are heavy.
When the power line voltage increases, B+ increases, so with the constant battery bias, the plate current goes up.
Battery bias voltage is in large increments for medium to high transconductance tubes.
1.5V bias change on a 5,500 micro-mho tube causes 8.25mA plate current change.
Last edited:
Hi 6a3
That's very useful information. Many thanks. I'm thinking fixed bias or battery bias.
Which did you prefer?
I usually use direct heated rectifiers for B+. That asks the question what to use for B-. Solid state would pretty much come up the same time as B+, maybe a trifle faster. It would be tempting to use a direct heated rectifier for the B-. In that case I'm supposing it would require an indirectly heated rectifier for the B+.
That's very useful information. Many thanks. I'm thinking fixed bias or battery bias.
Which did you prefer?
I usually use direct heated rectifiers for B+. That asks the question what to use for B-. Solid state would pretty much come up the same time as B+, maybe a trifle faster. It would be tempting to use a direct heated rectifier for the B-. In that case I'm supposing it would require an indirectly heated rectifier for the B+.
Would battery bias require bias to be turned off with B+ disconnected or is it OK to leave grid permanently negative even with no B+ present.
I assume that batteries must also be monitored and B+ switched off if they start losing voltage?
I assume that batteries must also be monitored and B+ switched off if they start losing voltage?
When used in the circuit and manner I stated, a group of series 9V Alkaline batteries will last for 2, 3, or several years. They all outlasted my changing from battery bias to another type of bias on several different amps of several different designs.
I always used Duracell Alkaline. 9V batteries are about 9.5V or so.
Years ago I built a battery tester that tested open circuit voltage, and internal impedance.
I compared 9V batteries from 3 major competitors. A friend of mine who also uses battery bias, independantly came to the same conclusion of which brand to use.
Your mileage might vary.
Yes, with the mains power off, take a look at the battery voltage at least once a year (use your floating DMM across the battery).
I never had a tube go bad or soft as a result of permanent negative grid bias.
A 1 volt bias drop on a 2A3 or 300B will change the plate current 5.5mA.
If you have 60mA, it would be 65.5mA.
I always used Duracell Alkaline. 9V batteries are about 9.5V or so.
Years ago I built a battery tester that tested open circuit voltage, and internal impedance.
I compared 9V batteries from 3 major competitors. A friend of mine who also uses battery bias, independantly came to the same conclusion of which brand to use.
Your mileage might vary.
Yes, with the mains power off, take a look at the battery voltage at least once a year (use your floating DMM across the battery).
I never had a tube go bad or soft as a result of permanent negative grid bias.
A 1 volt bias drop on a 2A3 or 300B will change the plate current 5.5mA.
If you have 60mA, it would be 65.5mA.
andyjevans,
For the fixed (adjustable) bias amp, I used an indirectly heated 5AR4 for the B+. The bias supply used a silicon diode. That was the conservative thing to do, but especially recommended if the output tubes are DHT.
I do not recommend battery bias for Push Pull output stages. An imbalance of plate currents of just a couple of mA can cause the bass to be distorted.
I have not used direct coupling between the driver and output tube(s).
I would strongly recommend using Self Bias for that kind of circuit.
For the fixed (adjustable) bias amp, I used an indirectly heated 5AR4 for the B+. The bias supply used a silicon diode. That was the conservative thing to do, but especially recommended if the output tubes are DHT.
I do not recommend battery bias for Push Pull output stages. An imbalance of plate currents of just a couple of mA can cause the bass to be distorted.
I have not used direct coupling between the driver and output tube(s).
I would strongly recommend using Self Bias for that kind of circuit.
andyjevans,
I forgot to mention, the battery bias amplifiers also were liked by those who heard them.
Sometimes I wonder, is it just that everybody is nice? No, I know that the various battery bias designs did sound good too.
Yes, there were various designs of amps with different tubes, different bias, different circuits, that I did not have others listen to, they were not good enough for my taste.
I may do some battery bias amps again.
It is all great fun to build, and enjoy listening to music.
I forgot to mention, the battery bias amplifiers also were liked by those who heard them.
Sometimes I wonder, is it just that everybody is nice? No, I know that the various battery bias designs did sound good too.
Yes, there were various designs of amps with different tubes, different bias, different circuits, that I did not have others listen to, they were not good enough for my taste.
I may do some battery bias amps again.
It is all great fun to build, and enjoy listening to music.
I can second Walter's recommendation of the JJ-2A3-40; a very nice tube. The lower filament voltage allows use of AC on the filament without hum issues.
Scott
- Status
- Not open for further replies.