Improved SCA-35/ST-35 Performance

for the DIY ST35, the Dynakit and every other PP EL84

Improved SCA-35/ST-35 Performance

Postby dcgillespie » Mon Oct 04, 2010 9:50 pm

For those of you who love these amplifiers (and their clones) as much as I do, I have posted a rather simple modification on the Tronola website (www.tronola.com), that significantly improves the performance of these units, and offers the documentation to back it up. It utilizes a unique new method of biasing I think you will find interesting. The article is written with plenty of how to information for the newer folks, and why explanation for the more experienced ones. If you are interested, go to the Tronola website and look for the section entitled Dave's Lab. There, you will find a link to take you directly to the article. I hope you will find it helpful in your quest for improved sound and performance.

Dave
dcgillespie
KT88
 
Posts: 399
Joined: Sun Mar 08, 2009 5:34 am
Location: Ball Ground, GA

Postby cedricb » Tue Oct 05, 2010 1:06 am

Are there any advantages to use this mod with the diytube version?
cedricb
 
Posts: 89
Joined: Mon Jan 28, 2008 4:49 am

Postby dcgillespie » Tue Oct 05, 2010 6:27 am

As long as you are using a Z565 transformer, or any of it's popular direct clone replacements (Magnaquest, Triode, etc.), the new EFB biasing arrangement is the only way to go, and will produce the same performance gains shown over any kind of existing cathode resistor bias system in use. Therefore, from that standpoint, it doesn't matter what the front end circuitry is, but obviously those who are using a diy board that includes the individual cathode resistor bias system built into it will need to modify their boards to allow for use of the new EFB regulator system. This should be rather easy to do. The new circuitry can be made to allow for individual adjustment of each tube or channel as discussed in the article -- just as the individual cathode resistor bias system does now. In any case, the new regulator circuitry could be wired point to point on a terminal strip, and mounted appropriately somewhere under the amplifier chassis.

For those using output transformers of 10K primary impedance (or greater) with screen taps around 50% (Heath used such specs in their UL 6BQ5 amps) and any form of cathode bias, use of the new bias arrangement will provide diminished returns, as transformers with those specs are optimized for traditional cathode resistor operation with 6BQ5 tubes. But for those currently using any of the Z565 "type" transformers and cathode resistor bias arrangements, the advantages of the new EFB bias system are very clear, and will provide the performance enhancements shown.

Dave
dcgillespie
KT88
 
Posts: 399
Joined: Sun Mar 08, 2009 5:34 am
Location: Ball Ground, GA

Postby Ty_Bower » Tue Oct 05, 2010 6:32 am

I've enjoyed reading your previous articles, and this one is no exception. Thanks for providing the link. There's a lot of interesting information in there, and I'll be absorbing it for a while.
"It's a different experience; the noise occlusion, crisp, clear sound, and defined powerful bass. Strong bass does not corrupt the higher frequencies, giving a very different overall feel of the sound, one that is, in my opinion, quite unique."
User avatar
Ty_Bower
KT88
 
Posts: 1494
Joined: Wed Mar 21, 2007 2:50 pm
Location: Newark, DE

Yeeeeaaahhhhh!!!!!!!!!!!!!!!!!!!!!!!!

Postby EWBrown » Tue Oct 05, 2010 7:08 pm

This will save a few steps of "navigation" to find it:

http://www.tronola.com/A_New_Look_At_An ... d_Rev0.pdf

PDF file, 18 pages, approx 1.4 MB.

You just know that I'm going to have to try this one... I really like its elegant simplicity, and use of readily available parts. (love)

Hmmm, this also sounds like fodder for a new Rev "E" design DIY35 PCboard. Shannon ????? ;) [:) I knopw that your original PC board vendor has faded away :(

I do have a factory built SCA-35 also in near-pristine condition, but I'll do this to an existing vintage ST-35, or my already seriously-hacked SCA35 first, as there isn't major board surgery required to implement this EFB circuitry Might as well leave the nice clean one unsathed for now...

I may have to obtain an LM337, if I don't have one or more stashed away in the parts collection.

As a former co-worker from Russia used to say: "GUDE STOFF!"

Since the EFB mod is all in the PA section, it should work nicely with my next (eventual) VA/LTPI / EL84s build using Shannon's ST-70 driver board, which I have already modified with resistor values changed for operation on ST35's available B+ voltages. So many amps, so little time... :'(

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby cedricb » Wed Oct 06, 2010 1:22 am

I'm currently building a revision D (I haven't received the transformers and the EL84 tube yet). So if somebody can decode the article and produce a nice step by step where to add the mod; that'll be great! :$
cedricb
 
Posts: 89
Joined: Mon Jan 28, 2008 4:49 am

Postby dcgillespie » Wed Oct 06, 2010 6:21 am

Thanks for the kind comments Ty -- I hope the concept helps everybody get the most out of their equipment. There are as many mods for Dynaco stuff out there as there are examples of their amps! What a testimony that is to the quality of their original product -- it never ceases to amaze me. This one is grounded though in solid reasoning and performance achievement, so I hope folks give it a try.

The simplicity of it was important to me too Ed, I've always appreciated that part of Dynaco's approach, and feel it was a huge part of their success. I would love to hear the results of your installation if you try it. You and Ty are both such huge contributors to this forum with proven capabilities, so I appreciate your interest in the concept.

BTW Ed -- I can't tell you how many times your general wit/humor/pics have given me a great start to the day -- after I pick myself up off the floor that is!

Dave
dcgillespie
KT88
 
Posts: 399
Joined: Sun Mar 08, 2009 5:34 am
Location: Ball Ground, GA

Postby DeathRex » Wed Oct 06, 2010 7:52 am

Could you use this on a SE design that has cathode biasing?
At first I wanted to be a ET. Now I are one.
User avatar
DeathRex
KT88
 
Posts: 1014
Joined: Wed Nov 18, 2009 8:13 pm
Location: Cortez, CO

Postby dcgillespie » Wed Oct 06, 2010 9:43 am

Death --

An EFB cathode regulator is ultimately a novel way to obtain fixed bias operation, without the use of a separate fixed bias supply. In addition, it's floating feature also has the unique ability to effecively eliminate any variations that a changing B+ imparts to performance (other than power output), which can have a significant positive impact on the distortion level generated in the output stage. Therefore, if you are working with a SE design where loading conditions are appropriate for normal fixed bias operation, using an EFB cathode regualtor would be a perfect way to go. It is simple, produces superb fixed bias operation, requires no bias supply to do so, and offers the majority of the benefits of regulated power supplies in the process. It is definitely applicable to any SE application requiring fixed bias conditions.

Dave
dcgillespie
KT88
 
Posts: 399
Joined: Sun Mar 08, 2009 5:34 am
Location: Ball Ground, GA

Postby DeathRex » Wed Oct 06, 2010 11:24 am

Well I'm building Mikael Abdellah's SE KT88 amp, except with Russian 6P3S-Es. I already have the 5KOhm output transformers. It seems the 6L6 will do 5KOhm in fixed and self biasing. I planning on running it UL at 350VDC and 50ma-55ma in. Unless I have a chart, I can't figure output impedance, but I can never go by a schematic 100%. If I can get the power out of fixed bias, using cathode bias, I'll be down with that.
At first I wanted to be a ET. Now I are one.
User avatar
DeathRex
KT88
 
Posts: 1014
Joined: Wed Nov 18, 2009 8:13 pm
Location: Cortez, CO

Postby Kyle K » Wed Oct 06, 2010 12:24 pm

6BQ5/EL84 data sheets list maximum grid resistor values of 1M for cathode bias and 300K for fixed bias operation. Would the 475K grid resistors used in the SCA-35 and ST-35 need to be reduced for this method of fixed bias?

Thanks,

Kyle
Kyle K
 
Posts: 29
Joined: Tue May 09, 2006 2:38 pm
Location: Mililani, HI

Postby EWBrown » Wed Oct 06, 2010 5:00 pm

Since it is still mostly cathode biased, with regulation tied through a resistive divider to the B+ voltage, the 470K resistors should still be OK.

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby dcgillespie » Wed Oct 06, 2010 6:20 pm

Kyle --

That's a great issue to point out. Since the bias point is fixed and independent of current flow through the tube, the same concerns for maximum grid DC resistance applies to operation with EFB as it does to traditional fixed bias designs.

The primary reason that the concern is very low here, is because one of the added benefits of using an EFB cathode regulator in these Dynaco amps is that the dissipation of each tube is now lowered significantly. Specifically, in my unit, the B+ to the OPT CT runs 380 vdc at quiescent condition. After the drop through the output transformer winding, it is no more than 375 vdc. The cathodes have about 15 vdc on them (all referencing ground), so that there is 360 vdc from plate to cathode. The tubes now idle at 27 ma each for the point of lowest distortion, which is made up of about 25 ma by the plate, and 2 ma by the screen. This means that the plates are now only dissipating 9 watts each, against a 12 watt Design Center rating, or even a 14.4 watt rating of the more modern Design Maximum rating system. The maximum grid DC resistance rating for a tube is based on operation at the maximum dissipation rating for the tube. Therefore, since the tubes are operating at least 25% below maximum ratings, the concern is very low here.

For those who are concerned about this issue, the 470K grid return resistors could be reduced to 220k, and the coupling caps increased to .22 mfd each. The added load this would place on the phase inverter stage would be insignificant for the drive level required.

Thanks for your interest.

Dave
dcgillespie
KT88
 
Posts: 399
Joined: Sun Mar 08, 2009 5:34 am
Location: Ball Ground, GA

Postby Kyle K » Thu Oct 07, 2010 1:01 pm

Thanks Dave and Ed. Lots of tube amp manufacturers, both past and present seem to ignore maximum grid resistor values as specified by the tube manufacturers, especially in fixed bias operation. As you say, the specified values are for maximum or design center dissipation, and they were probably conservative in setting the values. Thankfully, current production EL84's seem to be pretty uniform and well built compared to some other current production (or even NOS from the 80's) power tubes, so there would be less of a concern.

Kyle
Kyle K
 
Posts: 29
Joined: Tue May 09, 2006 2:38 pm
Location: Mililani, HI

Postby EWBrown » Thu Oct 07, 2010 7:10 pm

In my own designs I typically use 270K grid resistors for cathode bias ,and 100K for fixed bias, but with one major exception, which I recently discussed.

The 6DZ7, even though it is ostensibly two 6BQ5s sharing one bottle, really needs to have a lower value grid resistor to ensure stable operation.

I would SWAG that the 6DZ7 was designed to be used only in PP "Pentode" operation, or as two separate SE channels, , and that using it in PSE-UL mode never entered the equation during its development. Perhaps there is something unique to its internal elements which is different than simply being two 6BQ5 "room mates" other than the shared cathode and screen grids?

I initially tried 220K for both sections connected in parallel, but its operation was very unstable, and I eventually tried 150K, and then settled on 100K, which necessitated using larger (0.47 uF) coupling caps to compensate for the lower resistance driven load.

The 6CG7 SRPP driver stage has a fairly low output impedance, and plenty of drive current, so the lower resistance load is not an issue.

The additional 1K grid stoppers and 100 ohm screen stoppers are an excellent idea, and should be used in any similar design.

/ed B
Last edited by EWBrown on Fri Dec 03, 2010 6:33 pm, edited 2 times in total.
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Next

Return to stereo 35

Who is online

Users browsing this forum: No registered users and 3 guests

cron