• 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.

Dynaco ST35 power?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello my friends,

Anyone outthere who has meassured the full RMS power on the rev. Dynaco ST35 with one 470 resistor/470uF for each EL84 tube?
We came out as follows:

7,7 Watt RMS per channel at 1 kHz and 5 kHz , with a distortion of 4,75% close to break up the clean sinus. The draw of current per tube was approx. 37 mA. -3 dB at approx. 20 kHz. The original powertransformer delivered approx. 350 volt clean DC , almost without ripple, on each anode. The sound was really spectacular clear from the Z 565, but we could not reach even half of the wattpower mentioned from several netplaces???

After this experience we tryed with Mr. David Hafler´s single 100 Ohm /470uF resistor and cap. The tubes used came from Siemens , E84L rugged version, with only a minor difference at current draw and less than 10 mA per tube. Now the Dynaco ST35 showed up as follows:

approx. 12,5 Watt RMS at 1 and 5 kHz , distortion 4,15% - at 2 Watt and 5 Watt the distortion came as low as 0,12%/ 0,15%. Current draw per tube approx. 50 mA!! -3 dB point at approx. 31,7 kHz - Nice!!

It seems to us as if the REV. Dynaco ST35 has been better of with the old design from Dynaco/ Hafler instead of the single resistor design in the new design????

Try to look at :

www.tubeamp.hobbysider.dk and then click on :

www.diagrammer.hobbysider.dk

and find Dynaco project in the menu.

Anyone who experienced the same???

Best rgds
Kim Hansen
 
Hi Kim,

I've had a few of these pass thru my hands over the decades... still have a very nice one that's functional with a few upgrades. The original common cathode resistor is 95 ohms. If you wanted to split the output tubes for separates you should use 380 ohms (390 ohms is the standard value) per tube, not 470 ohms. Also, you did not mention plate voltage and/or if you altered the B+ supply in any manner as it used a 50 ohm resistor in the power supply pi filter. Were there any changes to the input/driver circuitry? Here's a link to the original schematic: http://www.triodeel.com/dynast35.gif

Can't recall my last specification check.... but I don't recall it being low on power output. The 7189 is a good output tube for these... pretty sure one of the last ones came with these installed actually and they have a higher voltage rating than the EL84/6BQ5.... and the ST-35 generally ran hot with higher voltages than spec'd.

Regards, KM
 
Dyna power

Hi. KM,

Thanks for your reply. I have changed the pi filter resistor from 50 Ohm to 1 Ohm because the 700uF used for the b+ has taken the top of the voltage to approx. 350 volt dc - original it is 360 volt dc. I have not used 12DW7 but one half ECC83 and ECC82 for each channel. We had to lower the rev. Dynaco double resistor from 470 Ohm as stated in the new design, to as low as 100 Ohm single resistor before we reached the 12,5 watt RMS. I know that the original resistor was the mentioned 95 Ohm. We could not her the lack of power from the 7,7 watt RMS, but the scope told another story with the single 100 Ohm resistor.

Thanks again for your answer.

best rgds
Kim
 
I haven't tried measuring for distortion, but I have checked my ST35 clone for peak output power and visible clipping on the oscilloscope. Unfortunately, I was not using a preamp during my testing. My maximum power output was limited by my CD player's output.

I achieved roughly 7.5 watts output, with no visible signs of clipping. This ST35 has original Z565 output transformers. The tubes are Reflektor 6p14p-eb, each with its own separate cathode resistor and 470uF cathode bypass capacitor. The resistor on each cathode was adjusted until each output tube was 35 mA at idle.

With a 20 Hz sine wave, an 8 ohm dummy load is connected to the 8 ohm tap. I'm reading about 22 volts peak to peak. Perhaps sometime I'll repeat the test with a preamp so I can drive the input to the ST35 hard enough to clip.

 
I would expect a little more single-tone output with a single cathode resistor rather than four, but I'm not sure it would be any louder on music peaks. I modified a Magnavox PP 6V6 amp to have separate cathode resistors with jacks to jumper the cathodes together - couldn't hear any difference. Didn't make any measurements, though. If output tubes were mismatched, separate resistors would be MUCH better.

40 mA x 350V is 14W of plate and screen power, so the tubes are at their limit. 50 mA will cook them pretty quickly.
 
Hello Ty Bower,

Thanks for your answer. I shall sleep better now after the statement from you. I bought the rest part of a SCA35 only because the Z565's and the powertransformer were still there . Tonight we got 12,5 Watt RMS out with a single 100 Ohm resistor, but the E84L Siemens took 53 mA each from the psu. The sound was good , fast and with fine definition - BUT - my lord the Siemens went hot. I think the best solution is to keep the tubes under 40 mA and accept the less power.

When I heard of 17 Watt per channel from the ST35, I thought that a mistake in the layout could be the reason. But now I am sure of the limit for the ST35.

Thanks - you have been a great help.

Regards
Kim
 
Hello Tom Bavis,

Thanks for your answer as well. No - I aggree with you. There is not a difference of + 3 dB from 7,7 Watt to 12,5 Watt , and it is for me impossible to hear the difference between the two designs. At the same time I think the distortion will make one turn down the volume long before the maximum has been reached. I will resolder the separate resistors for each tube - I would like my rather expensive Siemens tubes to last longer than a week.

Thank you very much for your reply.

The best rgds from

Kim - Denmark
 
Hi Kim,

With the ST-35, a more ideal solution would be to split the channels and have a common cathode resistor per channel, ie, pair the push-pull outputs together on a single cathode resistor... which would work out to 190 ohms... with standard values being either 180 or 200 ohms. That would be more logical to me.

Also, would suggest you try driving a single channel and see and what the power output is. IIRC, looking at the pics on your site, you're using some of the newer high capacity electrolytic capacitors... in some cases I've found those to actually consume some current (ie, they have high leakage) which steals it from the available supply to the amplifier circuitry which could account for less output power. Something to check.

My last suggestion would be to replace the 50 ohm resistor in the pi filter with a decent filter choke... obviously with a low DCR. Hum level was always a bit high with these units too.

Regards, KM
 
190 Ohm

Hi KM,
I would like to try this option. Today I bought 4 pcs. of 390 Ohm and it was easy to adjust each tube with the four 10 K pots to the accurate 35 mA per tube, and 345 volt DC on the plates.

If one could benefit from a single resistor of 190 Ohm per channel, this option has to be tested. May be the amplifier sounds even better with only one cathode resistor per pair of powertubes?

With the four resistor configuration we played "Amused to dead" by Roger Waters just to hear the quality and details from the ST35, and the amplifier played well and clear. We did not messure the output power and we were unable to detect any loss of effect.

Yes , I have used 4 pcs. of ROE 150uF/450v = 600uF plus one double can 50/50uF /550v cap from Jan Wüersten = 100uF. In all 700uF. I think the loss from the original 360v to 345v is depending on the high capasity and the leakage you mentioned. (if the main powertransformer is about the same for the SCA35 as for ST35?). I have bought the rest of an SCA35 for this ST35 clone and I have not any experience with the original ST35 at all. If the current floats away because of leakage, should the DC drop be even lower than the stated 345v?

I have used several resistors in the pi filter. For the moment 1 Ohm is in place, and the hum level is about - 80 dB messured. I don't know anything concerning the original spec's. I have a 2H/30 Ohm/2 A choke in stock - I could test this possibility as well.

Thanks for staying

Regards Kim
 
Hi Kim,

Yes, the SCA-35 uses identical iron to the ST-35.. have had a couple of these too over the years and still have at least one. The voltage drop seems high for increasing capacitance so I would expect cap leakage which would (partially) account for the lack of full power. Also think about the rectifier diodes you are using.... another possible source of voltage loss.

I would expect a single cathode bias resistor per pair and a good quality bypass cap would be the best scenario. Hammond has a good selection of chokes at reasonable prices... but check their DCR and ensure it's low. As for filter caps... maybe yank em and try some basic Sprague ATOM series...

Regards, KM
 
May be Sprague?

Hi KM,

May be the IN4007 have to be changed to another type with less loss?

I think the next test have to bee the 190 Ohm, we are not that far from the power wanted. - May be Sprague later?

I wonder whether the ohm difference on the primary windings of the Z565 could have influence on performance? If you messure from the +B red wire to the upper platewire (blue) the value in ohm, if I recall right, is lower than the same messurement from centerwire(red) to the lower platewire (blue/white). Typical a difference of approx. 20 Ohm . I have wired up the Z565 as stated by the rev. Dynaco ST35 design. It is possible to wire the Z565 up site down and still have some performance. just wondering .

Regards
Kim
 
The halves of the winding have the same number of turns, but the secon half is wound over the previous winding - so it's longer, and has more resistance. There are special winding configurations that wil give eual lengths, but Dynaco did not use them.

By using a single resistor per pair or quad, slightly more power is achieved, since the cathode voltage does not increase so much when driven out of class A (the resistor value is smaller - less drop). Large cathode bypass capacitors provide this too, but only for brief periods - enough for musical peaks. So peak power may be the same with separate cathode resistors, even though the continuous power capability is less.

I am not sure, but distortion may be lower with a common cathode resistor. The ability to exactly match cathode currents may be more important, though... it will reduce low-frequency distortion in the transformer to a minimum.
 
Wire hook up for Z565?

Hi Tom Bavis,

Yes, the same ohm difference has appeared on the Leak Stereo 20 outputtransformers, but compared to the Dynaco circuit Leak uses one ECC83 double triode to drive the output tubes , as where Dynaco use a single ECC82 triode. One signal wire from the plate and one signal wire from the cathode.

I have been thinking whether the impedance difference, by a single drive triode, would have any influence when driving the output tubes? Should the high ohm value on the Z565 primer winding side be connected to the output tube driven by cathode follower at the ECC82? - or the plate driver from the same tube?

It may not make any difference how the plate wires of the EL84 tubes are connected to the Z565, as long as the feedback is in right place.

Yes , it is not so important to get as much power as possible. The RMS value does not reflect the real peak power, and it is not sinus but music we need the amplifier for.

I will keep the 4 cathode resistors for a week or two . The sound is really fine within my personal top 5 amplifiers. No matter the power difference.

Regards Kim
 
Wrong text.

SY,

I apologize for writing wrong in the text:

"I have been thinking whether the impedance difference, by a single drive triode, would have any influence when driving the output tubes?"

Should have been:

"I have been thinking whether there is an impedance difference, by a single drive triode, and if so, this difference may have any influence when driving the output tubes?"

I understand from your message that the input impedance seen from the output tubes are about the same in spite driving a tube from the plate or the cathode?

I am not professionel engineer.

rgds. Kim
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.