I never liked the Dyna circuit, with one 1.56 Ohm resistor for two cathodes; That was a nice idea, when all the average person had was a 3% accurate Volt Ohmmeter.
You measured a 1.56V dry cell first, and then adjusted the bias until the test point made the needle move to the same point on the meter.
You had to use tubes that were specifically matched for the Dyna Stereo 70, at those exact quiescent operating points.
There was no way to tell if the pair was matched in that amp, or if with aging if they ever went unmatched.
Well, there was a method to tell if they were unmatched . . . Bass notes would distort, and would inter-modulate with the other musical notes. So much for an output transformer that had unmatched quiescent currents.
And the Global Negative Feedback did not fix that, it made the saturation worse.
No way to measure current matching, and no way to separately adjust the pair of currents (only one pot per output tube pair).
You probably are not using the B+ and filament power for a preamp (the Octal Socket on the front panel).
Why not wire in separate 1 Ohm 0.5% resistors, one for each cathode.
Tear out the wiring to the Octal socket (B+ and filament wiring), and provide access connections to the 4 new cathode resistors.
Check out the current matching.
If a new less expensive "matched" set of EL34 tubes does not sound good, check the voltage across each of the 1 Ohm 0.5% resistors.
Don't blame it on the tubes, until you have done that.
Oh, just for fun, install a matched set of KT77 tubes.
Modify slightly, as listed above.
Enjoy knowing that the currents are matched, instead of un-maatched
Some of those tubes sold as "matched" are not so well matched, Your Mileage May Vary (YMMV).
Matched tubes always match currents better with self bias resistors, versus using only one pot to adjust a pair of tubes.
I am just saying . . .
Listen, Listen, Listen.
Enjoy.
PS If I owned a Dyna Stereo 70, I would Reduce its output power, to tradeoff power for another advantage . . . this way
1. Tear out the bias circuits completely.
Install 4 self bias resistors, 4 bypass caps, and watch the currents of all those somewhat "matched" tubes have automatically quite closely matched currents.
Who believes the seller of "matched" EL34 tubes?
"Matched tubes" in your amp might have un-matched current?
Absolutely possible.
Most matched tubes are not at the same B+, DCRs of the plate taps and DCR of screen taps, filament voltage, etc.
Watch what happens to the cathode current, when your mains power goes up and down +3%, and -3%.
Some matched tubes are really that, others are not when installed in your amplifier.
One more thing. I loved my Dyna Stereo 70. But then I made it into a test bed for various stereo and various mono-block amplifiers.
I retained the Steel Chassis (I hate steel chassis, they transmit magnetic fields from power transformer and choke to the output transformer(s).
I also retained the Power Transformer.
The output transformers have gone to other amplifiers.
I am getting my money out of the old Dyna Stereo 70 amp, in enjoying designing, building, listening, changing to another topology, listening, etc.
You measured a 1.56V dry cell first, and then adjusted the bias until the test point made the needle move to the same point on the meter.
You had to use tubes that were specifically matched for the Dyna Stereo 70, at those exact quiescent operating points.
There was no way to tell if the pair was matched in that amp, or if with aging if they ever went unmatched.
Well, there was a method to tell if they were unmatched . . . Bass notes would distort, and would inter-modulate with the other musical notes. So much for an output transformer that had unmatched quiescent currents.
And the Global Negative Feedback did not fix that, it made the saturation worse.
No way to measure current matching, and no way to separately adjust the pair of currents (only one pot per output tube pair).
You probably are not using the B+ and filament power for a preamp (the Octal Socket on the front panel).
Why not wire in separate 1 Ohm 0.5% resistors, one for each cathode.
Tear out the wiring to the Octal socket (B+ and filament wiring), and provide access connections to the 4 new cathode resistors.
Check out the current matching.
If a new less expensive "matched" set of EL34 tubes does not sound good, check the voltage across each of the 1 Ohm 0.5% resistors.
Don't blame it on the tubes, until you have done that.
Oh, just for fun, install a matched set of KT77 tubes.
Modify slightly, as listed above.
Enjoy knowing that the currents are matched, instead of un-maatched
Some of those tubes sold as "matched" are not so well matched, Your Mileage May Vary (YMMV).
Matched tubes always match currents better with self bias resistors, versus using only one pot to adjust a pair of tubes.
I am just saying . . .
Listen, Listen, Listen.
Enjoy.
PS If I owned a Dyna Stereo 70, I would Reduce its output power, to tradeoff power for another advantage . . . this way
1. Tear out the bias circuits completely.
Install 4 self bias resistors, 4 bypass caps, and watch the currents of all those somewhat "matched" tubes have automatically quite closely matched currents.
Who believes the seller of "matched" EL34 tubes?
"Matched tubes" in your amp might have un-matched current?
Absolutely possible.
Most matched tubes are not at the same B+, DCRs of the plate taps and DCR of screen taps, filament voltage, etc.
Watch what happens to the cathode current, when your mains power goes up and down +3%, and -3%.
Some matched tubes are really that, others are not when installed in your amplifier.
One more thing. I loved my Dyna Stereo 70. But then I made it into a test bed for various stereo and various mono-block amplifiers.
I retained the Steel Chassis (I hate steel chassis, they transmit magnetic fields from power transformer and choke to the output transformer(s).
I also retained the Power Transformer.
The output transformers have gone to other amplifiers.
I am getting my money out of the old Dyna Stereo 70 amp, in enjoying designing, building, listening, changing to another topology, listening, etc.
Last edited:
You can check for tube match in an ST-70. It's fixed bias so all you have to do is pull a tube from one channel, get the reading from the one left in place, then replace it with the other one in that socket and get its reading. Compare the two, they will have had the same plate voltage and fixed bias voltage so the comparison will be valid. Do the other 2 tubes in that channel for uniformity. I believe Dynaco or maybe it was Acrosound that stated in their transformer catalogs that a 10% idle current mismatch was acceptable for two tubes on a PP tranny.
Definitely a first world problem.
Watch some world news & your 'problems' will seem insignificant.
When I first landed here many years ago, I struck up a conversation with one of the more prolific and knowledgeable speaker builders in the forums about amplifiers and speakers. The one thing he said to me that stuck all these years was eye opening and mind blowing...
"Yesterday's gold is today's silver."
"Yesterday's gold is today's silver."
20to20,
The only problem with pulling out one tube is that the B+ will go up (it is unregulated).
With higher B+, the screen voltage will go up, and the tube current will go up.
You are not testing the cathode current correctly versus the actual 4 quiescent currents with all EL34 tubes in place.
Did you ever adjust the bias to get the sense resistor voltage on one channel to be 1.56V, and then adjust the bias so the sense resistor read 1.56V on the other channel?
I did.
Now, you have to go back to the first channel, and its current sense voltage is different than 1.56V.
You have to go back and forth, back and forth, between the 2 channels, adjusting bias until Both are reading 1.56V.
That is one reason why I do not like that part of the circuit.
But, if you do not take that much attention to details, then just expect the performance to less good than it should be.
And, the input tube filaments are floating, with a 0.2uF cap to ground, they need to be raised by a resistive divider from B+, and have a bypass cap, so that the filament to cathode voltages do not exceed the input tube ratings. (both quiescent, and with peak signal voltage).
The Dyna Stereo 70 is not perfect. It works well, but a few simple changes make it so much more reliable, longer life tubes, and better performance.
Do not worry about replacing the output tubes.
The real Un-obtainum tubes are the 7199.
They do get destroyed, due to the poor floating voltage circuit topology. Two EL34 filaments and one 7199 filament are tied together. What is the filament to cathode leakage resistance of two EL34s (who knows). But I bet the "floating" filament secondary DC voltage is near 0 Volts.
That destroys the filament to cathode leakage over time, and the concertina becomes un-balanced.
I had a 7199 that even when cold, had 100k leakage resistance from the filament to the concertina cathode. Think of a 100k resistor in parallel with Rk, (100k through the 0.2uF cap to ground). That is a very poorly balanced concertina.
Check it out.
The only problem with pulling out one tube is that the B+ will go up (it is unregulated).
With higher B+, the screen voltage will go up, and the tube current will go up.
You are not testing the cathode current correctly versus the actual 4 quiescent currents with all EL34 tubes in place.
Did you ever adjust the bias to get the sense resistor voltage on one channel to be 1.56V, and then adjust the bias so the sense resistor read 1.56V on the other channel?
I did.
Now, you have to go back to the first channel, and its current sense voltage is different than 1.56V.
You have to go back and forth, back and forth, between the 2 channels, adjusting bias until Both are reading 1.56V.
That is one reason why I do not like that part of the circuit.
But, if you do not take that much attention to details, then just expect the performance to less good than it should be.
And, the input tube filaments are floating, with a 0.2uF cap to ground, they need to be raised by a resistive divider from B+, and have a bypass cap, so that the filament to cathode voltages do not exceed the input tube ratings. (both quiescent, and with peak signal voltage).
The Dyna Stereo 70 is not perfect. It works well, but a few simple changes make it so much more reliable, longer life tubes, and better performance.
Do not worry about replacing the output tubes.
The real Un-obtainum tubes are the 7199.
They do get destroyed, due to the poor floating voltage circuit topology. Two EL34 filaments and one 7199 filament are tied together. What is the filament to cathode leakage resistance of two EL34s (who knows). But I bet the "floating" filament secondary DC voltage is near 0 Volts.
That destroys the filament to cathode leakage over time, and the concertina becomes un-balanced.
I had a 7199 that even when cold, had 100k leakage resistance from the filament to the concertina cathode. Think of a 100k resistor in parallel with Rk, (100k through the 0.2uF cap to ground). That is a very poorly balanced concertina.
Check it out.
Last edited:
Don't worry about it, it works. You are only comparing the relative current one tube at a time. The slightly higher plate voltage isn't enough to create a problem. You are using the amp simply as a tester.
20to20,
It may, or may not, work well, depending on the particular output tubes.
However, the plate current does not change much when plate voltage varies.
The plate resistance, rp, is medium to high.
What will change the plate current more than a varying plate voltage, is when the screen voltage varies.
Look at plate current versus screen voltage, little change.
Look at plate current versus plate voltage, large change.
Those are worlds apart.
It may, or may not, work well, depending on the particular output tubes.
However, the plate current does not change much when plate voltage varies.
The plate resistance, rp, is medium to high.
What will change the plate current more than a varying plate voltage, is when the screen voltage varies.
Look at plate current versus screen voltage, little change.
Look at plate current versus plate voltage, large change.
Those are worlds apart.
Last edited:
The plate and screen voltage will not change. You use the same socket for each test and leave the other empty. You use the same OPT half-winding and screen tap. The starting voltages will be the same for each test.
The voltages remain at one level when 3 tubes are in place, no matter which of 2 sockets on one channel is empty. Each channel may have a different bias voltage at the required/recommended current, so treat each pair separately, not as a quad.
But, the voltages will change when 4 tubes are in place, including the screen voltage.
High transconductance EL34 tubes, with slightly different transconductance, and changing screen voltages is going to change the plate current (not always equally).
As I have observed, and as I have said before:
"All matched tubes are equal, just some matched tubes are more equal than others" -
Quote stolen and modified from "Animal Farm", by George Orwell.
But, the voltages will change when 4 tubes are in place, including the screen voltage.
High transconductance EL34 tubes, with slightly different transconductance, and changing screen voltages is going to change the plate current (not always equally).
As I have observed, and as I have said before:
"All matched tubes are equal, just some matched tubes are more equal than others" -
Quote stolen and modified from "Animal Farm", by George Orwell.
You use the same socket for uniformity of the test. The winding DCR is different from side to side on an A-470. The DCR is different between the CT and the screen tap and from tap to plate lead. So that's why it's best to use one socket for all the tubes. Yes the differences are small between sockets of the same channel but it's better scientific method to keep the test factors constant and uniform.
My valve amp is not quite in the same league as yours, but 50 plus years on, its original Brimar and Mullard valves are still going strong - at least to my ageing ears!
Enjoy your special valves now - life is short!
Enjoy your special valves now - life is short!
I agree with Goat Guy.
Nice Stradivarius illustration!
There are 2 solutions when it comes to a premium Hasselblad 6x6 camera, and Hasselblad premium lens.
1. Keep it stored away.
2. Use it how it was intended to be used, take it out, and take pictures.
Lots of digital camera owners love their 3 inch LCDs. The Hasselblad Analog Viewfinder is also 3 inches.
And there is a built in magnifier, if you use the waist level finder.
Thats right, 2-1/4 x 2-1/4 = 3.2 inches diagonal (just like the digital boys measure their LCDs . . . diagonally).
Wow!
Nobody seems willing to slightly modify a Sacred Dyna Stereo 70.
The value will go down from where it is now ($500, $700, or whatever a mint one is worth today).
I have modified amplifiers to include 10 Ohm resistors in each plate lead, with test points across the resistors, etc.
Or, it is easier to put 10 Ohm resistors in the cathodes, and measure from the cathode to ground.
I have to apologize for ruining perfectly good (well almost perfect) amplifiers.
Is it about looks, perfect original condition and parts, or how it sounds?
What about a 1928 Hupmobile car? Would you drive it if you had one in perfect condition?
Jay Leno enjoys driving his cars. Ask him.
Can we hear something from the Original Poster as to what he thinks?
Please?
Nice Stradivarius illustration!
There are 2 solutions when it comes to a premium Hasselblad 6x6 camera, and Hasselblad premium lens.
1. Keep it stored away.
2. Use it how it was intended to be used, take it out, and take pictures.
Lots of digital camera owners love their 3 inch LCDs. The Hasselblad Analog Viewfinder is also 3 inches.
And there is a built in magnifier, if you use the waist level finder.
Thats right, 2-1/4 x 2-1/4 = 3.2 inches diagonal (just like the digital boys measure their LCDs . . . diagonally).
Wow!
Nobody seems willing to slightly modify a Sacred Dyna Stereo 70.
The value will go down from where it is now ($500, $700, or whatever a mint one is worth today).
I have modified amplifiers to include 10 Ohm resistors in each plate lead, with test points across the resistors, etc.
Or, it is easier to put 10 Ohm resistors in the cathodes, and measure from the cathode to ground.
I have to apologize for ruining perfectly good (well almost perfect) amplifiers.
Is it about looks, perfect original condition and parts, or how it sounds?
What about a 1928 Hupmobile car? Would you drive it if you had one in perfect condition?
Jay Leno enjoys driving his cars. Ask him.
Can we hear something from the Original Poster as to what he thinks?
Please?
Last edited: