Topic: John's Build Pics

[jravacio]

good job as usual John.......

your voltages are spot on....you will find that when loaded those voltages will fall,

so it is better to start them a bit higher, than worry later because your voltages are low...

"gawang Cavite" is true.....

while working at Telefunken Semiconductors in the mid 70's
i had a co-worker from Bacoor who went to a trade school PCAT....

seeing how German production machines were made,
he made his amps look like they were german made......

he later went on to become testing manager after several years...

had i known i would have recommended that you use a voltage doubler as ht winding instead....
reason is so that you can use the half voltage available rather than deriving it from the main B+...
829b G2 needs 200 volts, i run them a bit lower at around 180 volts....you can tweak this voltage also...

Thanks Tony oh great you've worked at Telefunken Semiconductors back then and there were many electronics enthusiast and hobbyists. Unfortunately, there are very few electronics enthusiast and hobbyists now a days (well as far as I know).

I've worked for a company of over 700 employees. I only know 3 of us are doing electronics as a hobby. 2 of them are focused on firmware and hardware development so I guess I was left alone on the analog side of things.


Thanks for the suggestion, I'm considering it. I was actually contemplating on winding another transformer for this build using RM18 Cores and shelve this power transformer for some other builds.

My initial plan was to run the 829b in triode mode similar to the japanese designed schematic below. But I have my hesitations, because the G2 is way above the limit of 225VDC Max per datasheet. He ran the G2 at 316VDC. I don't have a datasheet of 829b with a triode connection in it.

I'm thinking of a Mosfet Regulated G2 Power Supply similar to what you've suggested for my 6V6 Push-Pull.

Any thoughts? I saw your upcomming 829b having 0A2 regulator tubes in it. Was that for the G2?



Reference:http://www.asahi-net.or.jp/~zh7y-tkyn/829B.htm
 

Very detailed DIY.

Im learning so much for these kind of details... Keep it coming...

Thanks Join us at elab on the link provided by Tony.

[tony]

My initial plan was to run the 829b in triode mode similar to the japanese designed schematic below. But I have my hesitations, because the G2 is way above the limit of 225VDC Max per datasheet. He ran the G2 at 316VDC. I don't have a datasheet of 829b with a triode connection in it.

I'm thinking of a Mosfet Regulated G2 Power Supply similar to what you've suggested for my 6V6 Push-Pull.

Any thoughts? I saw your upcomming 829b having 0A2 regulator tubes in it. Was that for the G2?



Reference:http://www.asahi-net.or.jp/~zh7y-tkyn/829B.htm
 

my attempts at a parallel triode mode for the 829b was a complete disaster...
but the 5894 was a complete success, the 5894 had a single cathode
making both halves equally matched as far as transconductance was concerned,
(there is only one cathode structure....)

G2 at 316 volts? i don't know.....
the 832 pp amp i built ran at 180 volts G2 with lots of filtering...
the Audoromy FU29 that i repaired had 200 volts on G2 and
is now well behaved according to the owner....
anyway running G2 at 316 volts only meant running higher G1 negative bias...
i see that he used combination of fixed and cathode bias in there,
i might do it also in my upcoming 829b pp amp build..
but overall if you want  the best reliability, then cathode bias is it...
the 2 0B2 gas reg tubes are used as voltage reference for the input stage and
yes, for the G2 supply as well.....a bit complicated perhaps...

btw, in the build article, he did mentioned that he dropped the G2 voltage later on,
probably realized that running the G2 as such high voltage only shortened tube life....

[jravacio]

my attempts at a parallel triode mode for the 829b was a complete disaster...
but the 5894 was a complete success, the 5894 had a single cathode
making both halves equally matched as far as transconductance was concerned,
(there is only one cathode structure....)

G2 at 316 volts? i don't know.....
the 832 pp amp i built ran at 180 volts G2 with lots of filtering...
the Audoromy FU29 that i repaired had 200 volts on G2 and
is now well behaved according to the owner....
anyway running G2 at 316 volts only meant running higher G1 negative bias...
i see that he used combination of fixed and cathode bias in there,
i might do it also in my upcoming 829b pp amp build..
but overall if you want  the best reliability, then cathode bias is it...
the 2 0B2 gas reg tubes are used as voltage reference for the input stage and
yes, for the G2 supply as well.....a bit complicated perhaps...

btw, in the build article, he did mentioned that he dropped the G2 voltage later on,
probably realized that running the G2 as such high voltage only shortened tube life....

Okay thanks for those information. It really helps mitigating possible issues.

The article was hard to understand and yes when google translate there was a mention of "drop the voltage of the SG". That would be the G2, but he never mentioned how many volts.

Regarding "running G2 at 316 volts only meant running higher G1 negative bias", that might explain why he had a separate 50VX2,0.1A Noguchi transformer inside the chasis for the fixed bias. The combination of fixed bias and cathode bias is also good to consider.

One last question, if I'll go regulating the G2 say 200VDC or below say 180VDC, I would like to have a safety feature in the case that the G2 voltage regulator goes on failure. The G2 would go floating and the potential will be the same as the Plate Voltage. This is the same as triode connection right? if my plate voltage is around 350V-400VDC then my tubes goes BOOM!

How do you handle such failure scenario? Thanks again.

[tony]

One last question, if I'll go regulating the G2 say 200VDC or below say 180VDC, I would like to have a safety feature in the case that the G2 voltage regulator goes on failure. The G2 would go floating and the potential will be the same as the Plate Voltage. This is the same as triode connection right? if my plate voltage is around 350V-400VDC then my tubes goes BOOM!

How do you handle such failure scenario? Thanks again.

remember that with tetrodes and pentodes the voltage at G2 controls
plate current unlike with triodes, where plate voltage on triodes
determine plate current, assuming grid1 voltage is held constant...

the thing you have to watch out for is the plate dissipation, make sure you do not
exceed 30 watts on either plate, i run them cooler at no more than 20 watts per plate..
adjust your tube bias accordingly...

[jravacio]

remember that with tetrodes and pentodes the voltage at G2 controls
plate current unlike with triodes, where plate voltage on triodes
determine plate current, assuming grid1 voltage is held constant...

the thing you have to watch out for is the plate dissipation, make sure you do not
exceed 30 watts on either plate, i run them cooler at no more than 20 watts per plate..
adjust your tube bias accordingly...

Sorry but I have not explained well the failure scenario.

Here's the scenario:

My 829b amp is running in Class AB, in pentode connection. The G2 is voltage regulated at 180 VDC and Plate Voltage is 350VDC. 829B are dissipating 20 watts per plate.

While the amp is running, G2 voltage is suddenly lost due to regulator failure or due to G2 wire loose connection. This will result into G2 floating and voltage goes up(probably equal to the plate voltage?). If I won't notice this, it will result into 829bs over-dissipating then dies.

(I've experienced this scenario on my tube tester with a 6L6 on it and I forget to set the G2 on the right settings(hence its floating) and the plate current goes up way above what I've expected. But since 6L6's G2 and Plate Voltage is close 450 and 500 VDC limits respectively. It was not a problem. For for the likes of 829b that has very low max G2 voltage, it might be.)

Is that scenario, something to be worried about? or I'm speculating some very low likelihood scenario?

For the meantime, I'm googling some answers

[jravacio]

Tony, after reading the article below I think a bleeder resistor from G2 to ground will prevent the G2 from floating in case of G2 voltage regulator failure. well unless, I've misunderstood the whole article.

http://www.photonisusa.com/wp-includes/pdf/TP122.pdf

[tony]

John, no voltage on G2 cuts off the tube....but leaving elements floating on any tube is no good...

[jravacio]

John, no voltage on G2 cuts off the tube....but leaving elements floating on any tube is no good...

Okay, I'll prepare a full schematic and post it at elab for review.

Thanks a lot for the tips. Really appreciate it.

[tony]

just like a triode that had an open grid leak,
it becomes a diode and acts like a diode,
only the circuit impedance will  limit plate current,
it may even red plate, but tubes are so forgiving of this kinds
of mistakes, just power down, fix the problem,
and the tube will work like nothing happened....

[jravacio]

just like a triode that had an open grid leak,
it becomes a diode and acts like a diode,
only the circuit impedance will  limit plate current,
it may even red plate, but tubes are so forgiving of this kinds
of mistakes, just power down, fix the problem,
and the tube will work like nothing happened....

Thanks Tony, it's very informative. Due to the intrinsic robustness of these tubes, it got me thinking that the fail safe I discussed above was actually a "Nice to have" rather than a necessity.

[jravacio]

A tour at AT&S factory



I'm a fan of AT&S transformers(as seen on the picture above, I have a few pairs of OPTs). Over the weekend I had the privilege of visiting Edrel's manufacturing center at EPZA Cavite.

Let us see what and how AT&S transformers are made and why they're world class.

1) Transformers are conservatively designed, that means they're not squeezing in more watts per kilogram. Effectively for power transformer they run cooler and for OPTs have greater frequency response

2) High-end magnetic wires are carefully chosen and used based on current density, thermal ratings, and other quality factors.

3) These are SET OPTs for 845 in the making. Yes, that's 24 interleaved layers of perfectly wound wires of AWESOMENESS.



4) Computer controlled winding machine for accurate number of turns. A wire tension tool is pulling the wires from the bobbin and the spool so that every turn has accurate tightness. Bifilar winding is done by hand instead of the tension tool.



5) Japanese Z11 EI Core. Yes, simply Japanese greatness.

6) High-end Insulating materials. Nomex, Teflon. etc. The works! These materials are quite pricey.



What's the other side of the tour?
Here are some M18 EI Core for some serious Power Transformer Winding.
All these quality parts came from Edrel, get yours now

[dana]

cool reporting on your AT&S  tour...

sayang, I could have dropped by to meet you too, neighbor ko lan yang si Great Edrel dto sa EPZA.
for sure, I treat tayo ni Great Edrel sa Shakeys nearby! Sideline na lan nya ngayon yan cguro...malakas raket nyan sa LED lighting ngayon
am sure, na meet mo rin  assistant nya, si Benji...pag may pyesa ka need na di makita ni Edrel, si Benji for sure may alam kung san nkatago

[jravacio]

cool reporting on your AT&S  tour...

sayang, I could have dropped by to meet you too, neighbor ko lan yang si Great Edrel dto sa EPZA.
for sure, I treat tayo ni Great Edrel sa Shakeys nearby! Sideline na lan nya ngayon yan cguro...malakas raket nyan sa LED lighting ngayon
am sure, na meet mo rin  assistant nya, si Benji...pag may pyesa ka need na di makita ni Edrel, si Benji for sure may alam kung san nkatago

Oi dana, malapit ka lng pala. Nasa Imus lang din ako. I'll hopefully meet you sometime. Edrel treat me as well.

hahha yes, Edrel's place was a wonderland of parts and stuff. Too many to buy them at once. hahaha:) I met Benji the first time last 2014 Hifi show too. He was there over the weekend explaining how things were done.

[tony]

John, those 845 OPT's are for a special audiophile......

[Courage]

Awesome.... Nomex and Teflon being used here are products of Company i work..

[dana]

John, those 845 OPT's are for a special audiophile......

probably a big shot celebrity audiophile?

[Tsnad]

Wow... everything is awesome... ganda ng mga trapo at yung mga project ni Sir John

[jravacio]

John, those 845 OPT's are for a special audiophile......

Yes like dana said looks like a big shot audiophile. I was hoping to see at least some pics once they're built or hear it if its for exhibit during the hifi show.

Awesome.... Nomex and Teflon being used here are products of Company i work..

Good to know

Wow... everything is awesome... ganda ng mga trapo at yung mga project ni Sir John

thanks heheheh

[tony]

probably a big shot celebrity audiophile?

yes, Edrel and I have been discussing since more than a year ago...

[jravacio]

829b Push-Pull Amplifier



Overview

The thing about DIY is that you can do it in your spare time. Yes, it took me about 7 months to finish this amplifier. But it was all worth it. I had doubts on my design, thinking it will not work the way I want it but it turned out to be awesome.

On top of the laws (Ohms, thermodynamics, etc.) that we have to follow, designing an amplifier has subjectivity as well. And the designer has to balance things or be biased on some aspects of it. I guess an analogy to this is a man looking for a woman. At some point he has to determine if he wants to have a high performance woman at a high cost or a woman with moderate performance at a moderate cost. Performance and cost are most likely to go together but in worst cases it could be a low performance at a high cost. John, what are you talking about? Amplifiers are our mistresses... Well that's what I was suspecting  hahahha

But anyway, this amp was loosely based on Seiichiro Suzuki 829b Push-Pull amplifier. However, instead of using a cathodyne phase inverter, I took the route of using LTP. Many amps like Audioromy FU29 have already used Cathodyne Phase Inverter. The main advantage of LTP over Cathodyne is that LTP is very linear. The 829b needs to be driven by low impedance output and this LTP using 12AU7 is a good candidate.

Playing safe with the input driver design, I have used a 12AT7 SRPP like you usually see on HIFI amps. Amplifiers using SRPP were called "distortion less audio amplifier" way back in the 50's.

The Output Stage uses the exquisite 829b a double tetrode tubes used in VHF radios in its hay day. I have been collecting a few of these tubes since about two years ago. And prices including the Russian GU-29 soared up on ebay after Cayin and Finale Audio made their amps using these tubes.

...And here's how I did it.

Power Transformer Winding









Enclosure layout, Drilling and Paint Job









Point-to-Point - Look mom, no PCB



Gut Shot


Cathode Bypass Capacitors


HV, Bias and Heater Supply



Back Panel


Side-by-side with my 6V6 Push-Pull amplifier



In Action


12AU7 Glowing


829B Glowing


Sound Review



Wifey's Review
Listened to mostly Crossover Jazz
This design is one of those that allows one to listen to all instruments/frequencies in a song. In fact, it was the first time I noticed a husky voice from a song I've been hearing forever from radio stations (okay, it would be a little too unfair to compare far too distant sound qualities, but...) But even if you are hearing these instruments and voices all at the same time, they do not compete with each other for your attention. What each does is do their part, and complement each other, like how a good sound engineer should arrange a song.

My Review

This amplifier sounds like a MOSFET in a tube. Clear, Punchy and Lively. I have listened mostly to 90's music to test this amp and turned out awesome. I could say this is not a Luscious kind of amplifier.



Acknowledgements

As always, this amp could not be made without all the advises from Tony Tecson. I am and the rest of the DIY community is thankful of having you around.

Edrel Sison for bringing the best goddies we could have for our builds.

Gerry Sta. Maria for the driver tubes.

Enrico Resurreccion for the wood works.

People at Elab and fb group "POWER AMPLIFIER & ELECTRONICS CIRCUIT TECHNICIAN FORUM" for continuous technical discussions

[tony]

congrats John, it is my lifelong passion to get you and the other younger people into tubes.....

now you have a lot of tricks in your toolbag, ready for use in any and all your tubes endeavor....

keep the fire in the bottle burning....

[jravacio]

Thanks again Tony...

I'm still contemplating what would be my next build. I have all the parts to build a KT88 push-pull and parts to complete an 832A push-pull amplifier. However, I could not get to the bottom of this hobby until I get to learn how to wound my own Output Transformers. My Z11 and RM18 cores are hiding in the boxes for months, waiting to get used. 

[kayee]

congrats John, it is my lifelong passion to get you and the other younger people into tubes.....

now you have a lot of tricks in your toolbag, ready for use in any and all your tubes endeavor....

keep the fire in the bottle burning....

Wow...nice build sir.

[tony]

kayee, musta na? mag pp amp ka na, kaya mo yan....

[kayee]

kayee, musta na? mag pp amp ka na, kaya mo yan....

sir tony  :)nag start pa lang ako sa enclosure ng gu50 ko na single ended sir paunti-unti muna....after nito mag push pull ako sir gisto ko rin subukan yung sq ng push pull.....

[jravacio]

Wow...nice build sir.

Thanks Kayee..Gawa ka din...join our technical discussions at elab http://www.elab.ph/forum/index.php?topic=34991.new;topicseen#new

[edrel sison]

Congratulations John! 7 months is such a short time for busy people like you and me!

[jravacio]

Congratulations John! 7 months is such a short time for busy people like you and me!

Thanks Edrel! time for a new project ulit heheheh

[tony]

Thanks Edrel! time for a new project ulit heheheh

John, yang si Edrel tatahi-tahimik pero high ender yan....nakita ko yung tube amp nya sa facebook...
Edrel makes world class OPT's for 845 SET amps for a bigtime audiophile....

[jravacio]

John, yang si Edrel tatahi-tahimik pero high ender yan....nakita ko yung tube amp nya sa facebook...
Edrel makes world class OPT's for 845 SET amps for a bigtime audiophile....

oo nga eh hehhe 300B at 10Y na preamp.

Sana mag e-exhibit yung  nagpagawa ng 845 OPTs sometime sa Hifi show.