• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

JJ 2A3-40 for Lynn Olson Amity

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello!
My friend has sold me all parts for building the Lynn Olson Amity, and it has the power transformer but not choke and output transformer.

Here is the circuit.

http://www.nutshellhifi.com/IT-Triode-Amp.gif

I have questions for you guys about substitute JJ 2A3-40 for the output tubes.

1) If I want to use the JJ 2A3-40, is the operating point (310v) still applied?

JJ Electronic - 2A3 - 40

2) As the suggested primary impedance of output transformer is between 6 to 8K, Which impedance should I go for?? 6.6K or 8K??

Many thanks!
 
JJ 2A3-40 seems to have a 300B electrical charateristics and physical form factor fitted with a 2A3 filament. It should be safe to opt for 450V @ 60mA bias. The kind of difference of OT impedance you are concerned with is perhaps secondary to winding technique, materials used and other differences in specifications.
 
JJ 2A3-40 seems to have a 300B electrical charateristics and physical form factor fitted with a 2A3 filament. It should be safe to opt for 450V @ 60mA bias. The kind of difference of OT impedance you are concerned with is perhaps secondary to winding technique, materials used and other differences in specifications.

Which of the following would you suggest for Amity with JJ 2A3-40??

TTG-EL34PP - Tube output transformer [6,6kOhm] 2xEL34 / 2x6L6 Push-pull or similar - Shop Toroidy.pl

TTG-EL84PP - Tube output transformer [8kOhm] 2xEL84 Push-pull or similar - Shop Toroidy.pl
 
Just to be clear: the JJ 2A3-40 is a 2A3 NOT a 300B. It biases up at the classic 2A3 operating point e.g., Ep= 250V, Eg= -45V, Ip= 60mA. And it follows the plate curves for a 2A3. The difference is that the 300B plate allows for a much more aggressive operating point. I have used 300V @ 75mA with no issues.
 
Just to be clear: the JJ 2A3-40 is a 2A3 NOT a 300B. It biases up at the classic 2A3 operating point e.g., Ep= 250V, Eg= -45V, Ip= 60mA. And it follows the plate curves for a 2A3. The difference is that the 300B plate allows for a much more aggressive operating point. I have used 300V @ 75mA with no issues.

So I should operate the JJ 2A3-40 at 310v as the circuit diagram indicated that is supposed to have conventional 2A3, right?
 
Interestingly, it sounds much closer to 300B than to 2A3.

Not convinced that a toroidal output is a great idea.

The toroid opt doesn't cost arms and legs for a trial, so why not?

Also, I have listened one of the finest 300B pp amplifier back in early 90's that is called LA Audio from Denmark, and it has toroid irons.

A bad EI or C type opt can sound bad.
 
I have JJ 300B tubes, and I also have JJ 2A3-40 tubes.

The plates look identical (size, shape, color).
The glass looks identical (size, shape).
The 2A3-40 has a black header that holds the pins (painted black).
The 300B has a white header that holds the pons.
The bottom of both 2A3-40 and 300B headers are white.

The 300B 5V filament connections are from far end to far end
(and the center is un-connected).
The 2A3-40 2.5V filament connects the far ends together, the other connection is to the center.

I would bet the filaments are identical except for the connections.
2.5V @ 2.5A, and 5V @ ~ 1.25A

The JJ 300B is a Mono-Plate tube.
The JJ 2A3-40 is a Mono-Plate tube.
These are both 40 Watt maximum plate dissipation.

What I can not tell is if the grids are the same or different on the JJ 2A3-40 and JJ 300B.
I bet they are the same, but I am not going to take excellent, well made, working tubes apart just to satisfy anyone's curiosity.
The JJ 2A3-40 and JJ 300B sound good.

There are other 2A3 Mono-Plate tubes, but they are either very old original RCA, or are made by some newer tube manufacturers (Electro Harmonix and Sovtek are examples).
But all of these are 15 Watt maximum plate dissipation.

There are very old stock and new manufactured dual plate 2A3 tubes. Valve Art is an example of a newer manufactured 2A3. But these also have 15 Watt maximum plate dissipation.

As to how a 2A3, 2A3-40, or 300B sounds in your system, I leave that to you.
You can run your own scientific, or non-scientific listening tests.
 
I have JJ 300B tubes, and I also have JJ 2A3-40 tubes.

The plates look identical (size, shape, color).
The glass looks identical (size, shape).
The 2A3-40 has a black header that holds the pins (painted black).
The 300B has a white header that holds the pons.
The bottom of both 2A3-40 and 300B headers are white.

The 300B 5V filament connections are from far end to far end
(and the center is un-connected).
The 2A3-40 2.5V filament connects the far ends together, the other connection is to the center.

I would bet the filaments are identical except for the connections.
2.5V @ 2.5A, and 5V @ ~ 1.25A

The JJ 300B is a Mono-Plate tube.
The JJ 2A3-40 is a Mono-Plate tube.
These are both 40 Watt maximum plate dissipation.

What I can not tell is if the grids are the same or different on the JJ 2A3-40 and JJ 300B.
I bet they are the same, but I am not going to take excellent, well made, working tubes apart just to satisfy anyone's curiosity.
The JJ 2A3-40 and JJ 300B sound good.

There are other 2A3 Mono-Plate tubes, but they are either very old original RCA, or are made by some newer tube manufacturers (Electro Harmonix and Sovtek are examples).
But all of these are 15 Watt maximum plate dissipation.

There are very old stock and new manufactured dual plate 2A3 tubes. Valve Art is an example of a newer manufactured 2A3. But these also have 15 Watt maximum plate dissipation.

As to how a 2A3, 2A3-40, or 300B sounds in your system, I leave that to you.
You can run your own scientific, or non-scientific listening tests.

Refer to the data of jj 2A3, it is capable 7-8w single ended power even operating at 300v, right?
 
I have JJ 300B tubes, and I also have JJ 2A3-40 tubes.

The plates look identical (size, shape, color).
The glass looks identical (size, shape).
The 2A3-40 has a black header that holds the pins (painted black).
The 300B has a white header that holds the pons.
The bottom of both 2A3-40 and 300B headers are white.

The 300B 5V filament connections are from far end to far end
(and the center is un-connected).
The 2A3-40 2.5V filament connects the far ends together, the other connection is to the center.

I would bet the filaments are identical except for the connections.
2.5V @ 2.5A, and 5V @ ~ 1.25A

The JJ 300B is a Mono-Plate tube.
The JJ 2A3-40 is a Mono-Plate tube.
These are both 40 Watt maximum plate dissipation.

What I can not tell is if the grids are the same or different on the JJ 2A3-40 and JJ 300B.
I bet they are the same, but I am not going to take excellent, well made, working tubes apart just to satisfy anyone's curiosity.
The JJ 2A3-40 and JJ 300B sound good.

There are other 2A3 Mono-Plate tubes, but they are either very old original RCA, or are made by some newer tube manufacturers (Electro Harmonix and Sovtek are examples).
But all of these are 15 Watt maximum plate dissipation.

There are very old stock and new manufactured dual plate 2A3 tubes. Valve Art is an example of a newer manufactured 2A3. But these also have 15 Watt maximum plate dissipation.

As to how a 2A3, 2A3-40, or 300B sounds in your system, I leave that to you.
You can run your own scientific, or non-scientific listening tests.

very good observations, and why not? ease of manufacture, can get two type birds in one go...just configure the filaments to satisfy either 2a3 and 300b...very clever manufacture...:cool:
 
Tubenstein,

How well will your 2A3 plate currents be matched?

Will you use a 1 Ohm or 10 Ohm 'sense' resistor in series with each 2A3 plate and test points in order to measure the voltage drop with a DMM and calculate the current, and current match?
"Matched" tubes are not necessarily matched at the exact plate to filament voltage, plate current, and grid bias voltage, that you will use them in your circuit.

I do not see a Toroidy specification of how much the push and pull DC currents can be different (un-balanced).

As to Lynn Olson's use of Air Gapped transformers, it is possible he used air gapped push pull transformers to change the shape of the hysteresis curve of the laminations.
I believe air gapping changes that curve at small signal.
And it certainly increases the amount of current before the laminations saturate.

I would bet that Lynn used a custom transformer built by someone like Bud Purvine.
 
Tubenstein,

How well will your 2A3 plate currents be matched?

Will you use a 1 Ohm or 10 Ohm 'sense' resistor in series with each 2A3 plate and test points in order to measure the voltage drop with a DMM and calculate the current, and current match?
"Matched" tubes are not necessarily matched at the exact plate to filament voltage, plate current, and grid bias voltage, that you will use them in your circuit.

I do not see a Toroidy specification of how much the push and pull DC currents can be different (un-balanced).

As to Lynn Olson's use of Air Gapped transformers, it is possible he used air gapped push pull transformers to change the shape of the hysteresis curve of the laminations.
I believe air gapping changes that curve at small signal.
And it certainly increases the amount of current before the laminations saturate.

I would bet that Lynn used a custom transformer built by someone like Bud Purvine.

I will use this module for each Amity monoblock along with the toroidy opt.

Tube bias control

I know he uses o-netics opt but too expensive.
 
Tubenstein,

That auto bias PCB you linked to, should work.
Using separate bias for each 2A3 will assure the current match.

When I do push pull amps, I use well matched tubes, and individual self bias resistors and bypass caps for each tube.
That makes it easy to check the real current match with new tubes in my circuit, and as the tubes age, if they are still conducting the same current.

Even a push pull pair of the old style 2A3 tubes that are only rated at 15W plate dissipation can produce 15 Watts of output power. Check the RCA tube manuals.
The JJ 2A3-40 should be able to produce more output power than that. But to do that, the operating voltage, current, and OPT primary plate to plate impedance are going to need to be set closer to the values of a 300B push pull amp.

Oh . . . Please let us know how it sounds when you get it completed.
 
Last edited:
I am late to this thread, but I do have a few comments to offer :^)

1) The 2A3 and 300B are very similar - gain, plate resistance, and transconductance are within 10-15% of each other. This is within the variation from sample to sample of the tubes. The circuit has voltage specs for both 2A3 and 300B; the limitation is the tube's heat dissipation limits. The operating points, normalized by the voltage, are very similar - the (shared) bias resistors are different in inverse proportion to the gain (mu) of the tube. So something like the 2A3-40 can be used at any voltage from 300v to 450v without any other changes. Note that the quiescent currents are on the high side - 85mA per tube for the 300B. If you use a "normal" P-P bias current, you will NOT be in the kind of deep Class A that is at the heart of this design.

2) The filament/cathodes are connected together, presumably to keep the bias resistor and its bypass cap out of the output current loop. But in spite of the requirement for "factory-matched" tube pairs, there will still be some difference in quiescent current, which will saturate the output transformer unless it has an adequate airgap. Note that an airgap capable of 20-60mA unbalanced current as specified is quite unusual, you may have to have one made with a custom airgap. Normally toroids have no airgap and are especially sensitive to this issue. (IIRC there are some, made for audio, that have sawed the toroid in half and re-assembled it to create an air gap.)

3) If you intend to build one, it is worth the time and effort to read Lynn Olson's description of the design: The Amity, Raven, and Aurora
His main point is to use the tubes in deep class A.That is why the output transformer impedance is so high. Normal push-pull, as specified in the 2A3 and 300B data sheets, uses an output transformer impedance that is about 2/3 of the single ended value, per tube - e.g. 4Kohms plate to plate (2K per tube) vs 3K for a single ended. He is using 3/2 (8K, 4K per tube). This means this P-P amp will make about 50% more power than a SET, instead of 200% more which is typical.
 
One great characteristic of staying in the class A region of a push pull triode amplifier is the damping factor is relatively constant throughout the complete class A signal swing.
This is true, even if there is no global negative feedback.
And because the quiescent current in the Amity is high in each output tube, the plate resistance is quite low.

The loudspeaker is dampened by the same amount in each direction, so there is no
"Woofer Walk", where the woofer peak to peak excursion is biased in one direction.
Woofer Walk is a bad thing, and the recovery after a transient is more difficult.

Now since both triodes are on all the time in class A, the tubes are effectively in parallel (two 300B, each with 700 Ohm plate resistance essentially in parallel = 350 Ohms).
That 350 Ohms is essentially driving 2k when the plate currents are still high, and in the class A region.
2k / 350 Ohms is a good number for good damping, and for low distortion.
But you get all that, plus the other good characteristics of push pull.

I remember being around (and working with) Lynn Olson. Those were good days.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.