Grid driven (ETM) KT88 SE amp design - final works on paper - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 28th May 2004, 12:09 PM   #1
Vortex is offline Vortex  Hungary
diyAudio Member
 
Vortex's Avatar
 
Join Date: Jan 2004
Location: Debrecen, Hungary
Send a message via ICQ to Vortex Send a message via MSN to Vortex
Question Grid driven (ETM) KT88 SE amp design - final works on paper

Hi everybody,


recently (abot 2-3 months ago) I started a thread here with a monoblock design, each block built with 1x E88CC + 1x ECC99 double triodes and with 1x KT88 end pentode.

This was a modified EAR-859 design with variable positive/negative feedback at the end, Class-A operation and KT88 grid-driven (also called ETM=Enhanced Triode Mode by Tim de Paravicini, original creator of the EAR-859).


Well, everything just got fine, lots of you here made proposals and gave critics about the design and the circuitry got ready finally.

Now, just to remember the design, each monoblock will be fully separated from the other one, so different chassis and own toroidal power supply. Each of them will operate at about 110mA and 320V anode voltage. Finally - according to my father - the whole design will give me about 10-13 Watts of pure Class-A SE power.

Okay.


Now, my most important question:

Circuitry finished, everything's okay on paper. Now we're at the point where we order the parts (tubes already bought).

My question relates to the power supply unit, especially to the two toroidal transformers.

My father ordered to pieces of it (for the 2 monoblocks), each rated at 75 Watts of power capability.

What do you think, will it be just enough for a nice wonderful sound, or is it still too small ?

I've been on another forums as well, where I was told, power supplies are the better the more oversized (until a level where it doesn't have any sense anymore of course).

Father told me, it's already oversized for my design. Now the question:

1) what do you think ?

Lots of you here have a lot of experience building single ended Class-A amps.

Let's say, the circuitry is ideal, just ultra-perfect (this isn't the case of course), you know what it should do in quality ...

2) what kind of power supply transformer would you pick for such a design / power ?

Some more information, perhaps it helps (or senseless here, dunno):

OT-s will be EI-core ones, heating will be stabilized DC, main power supply unit will also be stabilized.

3) Does it have any sense operating the heating for the tubes from a separate supply and small toroidal power transformers, or just leave it the old-fashioned way, like one power supply unit delivers all AC and DC power to everything in the system..?

One more thing: the monoblocks (each) will have 2 power-on switches. First I turn on the heating and then after waiting for some time (2-3 Minutes) I'll turn on the main unit. No delay circuitry. (I think it makes my blocks a bit unique, I decided to design them this way not because of quality reasons but because of "just for fun").


So, as you see, 3 questions came together finally.

Please help with advice if you can. Size of the blocks doesn't matter. Everything should be for high sound quality.

I'm confused at this 75 Watts for the mains - but who knows? Maybe father has right, maybe others!

Let's !
  Reply With Quote
Old 28th May 2004, 07:19 PM   #2
Vortex is offline Vortex  Hungary
diyAudio Member
 
Vortex's Avatar
 
Join Date: Jan 2004
Location: Debrecen, Hungary
Send a message via ICQ to Vortex Send a message via MSN to Vortex
Nobody ?

  Reply With Quote
Old 28th May 2004, 08:31 PM   #3
G is offline G  United States
diyAudio Member
 
G's Avatar
 
Join Date: Apr 2002
Location: Belleville, IL.
I always use transformers rated 2 times what the expected total current for the circuit would be. I normally use standard Hammond 300 series power transformers so I'm not familiar with using toroidals. What brand of transformers do you have in mind? What is your budget? What type of chassis are you going to use? Perhaps you should post a schematic.
__________________
Gavin
  Reply With Quote
Old 28th May 2004, 10:00 PM   #4
diyAudio Senior Member
 
fdegrove's Avatar
 
Join Date: Aug 2002
Location: Belgium
Hi,

First of all this is a class A design so the powertransformer needs only to be what the circuit demands on powerconsumption + a small margin.
If 75VA is already more than enough then I'd stick with that.

However, why toroidal xformers?

They're far too wideband and a simple EI xformer will filter out (acctually it just won't let it pass) RFI crud riding on the mains.

I'd recommend an electrotatic screen between primary and secondary and use a separate xformer form the B+ and the heaters to avoid capacitive coupling between the two.

That's really all there's to it....

Cheers,
__________________
Frank
  Reply With Quote
Old 28th May 2004, 11:06 PM   #5
Vortex is offline Vortex  Hungary
diyAudio Member
 
Vortex's Avatar
 
Join Date: Jan 2004
Location: Debrecen, Hungary
Send a message via ICQ to Vortex Send a message via MSN to Vortex
fdegrove, thanks for the suggestion.

Yes, it's already oversized with 75 VA. Then we won't modify it.

We chose toroidals because of space issues and the some better magnetic properties (toroidals told to be some "faster" or what, than normal ones, although I don't know, if it matters at all - probably not).

G, schematics will follow tomorrow or on Sunday.

Mark this topic, if you're interested - both of you. I'll be happy anyway if you look in here sometimes.


Chassis will be pure copper or aluminium, so non-magnetic and still electrostatic shielding. We'll then screw onto the sides and the top the wooden finish (from inside of course), the whole design will be rounded then and have a look just as the whole case would be made of wood.

Final look will either be some lacquered mahagoni color and pattern or "just" simple snow-white piano finish with gold-plated switches.

Buttons none at all, just 2 power-on switches and a gold plated top metal plate with my father's sign on it (laser engraved if I'm right here .. sorry for my English).

Somewhat that.

Preamp will be the same look, one unit, DACT 2-channel switched attenuator and DACT input selector. On the rear side some RCA plugs - that's it. Fully passive, just simple pass-through except the Phono stage. It will be the MC/MM design of Mr. Borbely with 2sk369 jfets at the MC level as the beginning - the rest you probably know (tubes). I'm also going to build in a DAC, I found a 24-bit DIY one - we'll see, how it performs.

The interesting thing will be the connection between preamp and the monoblocks, since I'm not going to use ultra-high quality interconnect cables and silver things with RCA plugs.. I'll just simply use BNC connectors on both sides (preamp-out/amp-in) with correct matched impedance. Some kind of better quality coaxial cable will fit.
(Inputs on the preamp will still be normal RCA plugs just to make sure I can change my sources at any time to anything else).

Still any ideas ? They're welcome.
  Reply With Quote
Old 28th May 2004, 11:25 PM   #6
diyAudio Senior Member
 
fdegrove's Avatar
 
Join Date: Aug 2002
Location: Belgium
Hi,

Quote:
We chose toroidals because of space issues and the some better magnetic properties (toroidals told to be some "faster" or what, than normal ones, although I don't know, if it matters at all - probably not).
Better magnetic properties do not necessarilly a better transformer make...quite to the contrary.

What is probably meant is that toroidal xformers have a smaller magnetic field.
This is really not a problem as with proper orientaion and layout inductive coupling is pretty much non-existant.

Faster? What is a fast powerxformer?
As said this is a class A design so current draw is pretty much constant.
If you're using low ESR quality PS caps this shouldn't be any issue at all...Provided it ever is anyway...

Happy building,
__________________
Frank
  Reply With Quote
Old 29th May 2004, 12:48 AM   #7
G is offline G  United States
diyAudio Member
 
G's Avatar
 
Join Date: Apr 2002
Location: Belleville, IL.
Quote:
Originally posted by Vortex
fdegrove, thanks for the suggestion.

Yes, it's already oversized with 75 VA. Then we won't modify it.

We chose toroidals because of space issues and the some better magnetic properties (toroidals told to be some "faster" or what, than normal ones, although I don't know, if it matters at all - probably not).

G, schematics will follow tomorrow or on Sunday.

Mark this topic, if you're interested - both of you. I'll be happy anyway if you look in here sometimes.


Chassis will be pure copper or aluminium, so non-magnetic and still electrostatic shielding. We'll then screw onto the sides and the top the wooden finish (from inside of course), the whole design will be rounded then and have a look just as the whole case would be made of wood.

Final look will either be some lacquered mahagoni color and pattern or "just" simple snow-white piano finish with gold-plated switches.

Buttons none at all, just 2 power-on switches and a gold plated top metal plate with my father's sign on it (laser engraved if I'm right here .. sorry for my English).

Somewhat that.

Preamp will be the same look, one unit, DACT 2-channel switched attenuator and DACT input selector. On the rear side some RCA plugs - that's it. Fully passive, just simple pass-through except the Phono stage. It will be the MC/MM design of Mr. Borbely with 2sk369 jfets at the MC level as the beginning - the rest you probably know (tubes). I'm also going to build in a DAC, I found a 24-bit DIY one - we'll see, how it performs.

The interesting thing will be the connection between preamp and the monoblocks, since I'm not going to use ultra-high quality interconnect cables and silver things with RCA plugs.. I'll just simply use BNC connectors on both sides (preamp-out/amp-in) with correct matched impedance. Some kind of better quality coaxial cable will fit.
(Inputs on the preamp will still be normal RCA plugs just to make sure I can change my sources at any time to anything else).

Still any ideas ? They're welcome.

Sounds like a doable yet very ambitious project. Please do keep us updated and post a schematic so that if any problems arise we will have a reference when trying to offer assistance, if any is needed. I assume you are using a ladder type attenuator for the passive pre that you are building?
__________________
Gavin
  Reply With Quote
Old 29th May 2004, 11:23 AM   #8
Vortex is offline Vortex  Hungary
diyAudio Member
 
Vortex's Avatar
 
Join Date: Jan 2004
Location: Debrecen, Hungary
Send a message via ICQ to Vortex Send a message via MSN to Vortex
Quote:
Sounds like a doable yet very ambitious project. Please do keep us updated and post a schematic so that if any problems arise we will have a reference when trying to offer assistance, if any is needed. I assume you are using a ladder type attenuator for the passive pre that you are building?
Attenuator will be ladder type, yes. Click here - I'll buy this type with 2 decks.

Actually, schematics are already okay. Why I'm not givin' it here for free?

Because I made some mistakes in it.

The principle was at the beginning, that my father - a very experienced electrician - was taking the EAR-859 as a starting point. Since we don't have it's tubes here in Europe and the KT88 performs also better (it's also told to be the ultimate among pentodes, just like the 300B or 2A3 triodes) .. well, he made some changes to the original EAR schematics to be able to use the E88CC (military ECC88) and ECC99 small signal triodes, just as the KT88 - all made by JJ Electronics .

Well, I didn't want a stereo amp, AND of course I told him that I'd like to have a separate preamp module. So he took off the switches, too.

He added just one more feedback - he doesn't believes to rumors and false information that feedbacks "kill the sound" and whatever - you know what I'm talking about. We recently also read an article about the "mighty" no-feedback SE amps and the advantages and disadvantages of NFB. Well, even in those ultra high-end "no feedback at all"-claimed Class-A designs, there ARE feedbacks of course, just a bit hidden from the eyes. So he just applied a feedback circuitry to the already modified schematics with a variable resistor at the end. With it's help we can set the feedback to either positive, zero, or negative. We'll then see, which feedback method sounds the best. The advantage of this will be, that each speaker configuration can be exactly matched to the amp. At the beginning I'll use my big 8 Ohm 3-way boxes with 12" bass drivers (lol) - well, it will sound, not too loud, but anyway.. this set will be happy of a bit more damping factor and spekaker control, not like a high-end stiff paper-cone 8-10" fullrange speaker. (I think you understand what I mean).

So that's it - no other modifications to the original EAR design. Just a feedback and other tubes - that's it.

I told my father, I really want quality and long lasting sound.. I don't want to play around changing my KT88's every half a year or so. This ETM design gives us more than 20,000 hours of ON-time (according to the EAR-859 specification and the KT88 may be even better, than EAR's tubes) .. we'll see. He also lowered the output power of the KT88 tubes, so they are by far below their limiting factors - but still at a good performing point.

Coaxial cable between preamp and monoblocks will also be used after an article's idea, where I read some interesting things: RCA plugs have an impedance of 100-150 Ohms. And we are still using it with who knows how much impedance-rated cables. At digital signal transfers, when impedance isn't matched, quadratic signal pieces will be either cut off at the beginning or end, edges rounded down, or it will have amplitude peaks at the sign which isn't very good at all I think. On normal audio frequencies it doesn't take much in account, but anyway, I'll stick to it, since the article (perhaps also a Borbely one??? I don't know) showed, how much better such a BNC-connector with conventional coaxial cable performs than a quite expensive all-silver cable at about $300 with RCA-plugs design and unmatched impedance. We'll see.

But let me go back to the point, why I'm not posting schematics yet: father saw a good idea in the Borbely MC preamp: almost every capacitor had it's parallel-connected pair with the original's 5% value. About 5% was it. Father told me it's important for crisp clear highs, because even mid-sized and larger caps have a slow reaction on high frequencies, they produce a lush warm too soft sound, so as I saw, he liked this idea and told me to parallel-connect small capacitors valued at 5% of the original's to each one in the ready-made design.

Well, I did it.

When he saw the circuitry, he just laughed up loudly I was doubling the caps everywhere where I saw one.

He already put in some smaller caps next to big ones, and I put even smaller tiny caps to those, too.

So, the actual design where we're at the moment, works, no problem, but fully unnecessary, he told me.

So I'll just ask him, which one to take out from the design, and I'll post the whole schematics it in the coming days.


Power supply for the blocks will be made as separate unit. I'm going to bring the DC power into the blocks via standard 3-pin computer power cables. Just to avoid interference between supply and OT / other parts. Perhaps an overkilling idea - dunno. We'll do it that way.

Power supply will be made with standard Graetz-Bridge, probably Hexfreds diode rectifying and then 100,000uF buffer-electrolytics with smaller ceramic or film ones in parallel for fast response. We'll stabilize both the heater DC and mains DC voltages - that's it. It will be quite simple, I think.


Ideas critics etc. still welcome - for some days from now on.

(I hope, my links work. )

Stay tuned
  Reply With Quote
Old 29th May 2004, 11:27 AM   #9
Vortex is offline Vortex  Hungary
diyAudio Member
 
Vortex's Avatar
 
Join Date: Jan 2004
Location: Debrecen, Hungary
Send a message via ICQ to Vortex Send a message via MSN to Vortex
yeah, the JJ link is bad.

Here's the correct one.
  Reply With Quote
Old 29th May 2004, 04:35 PM   #10
benny is offline benny  Australia
diyAudio Member
 
benny's Avatar
 
Join Date: May 2004
Location: Melbourne, Australia
Send a message via MSN to benny
building seperate PS to the main chassis is a bit overkill i think, but not a really bad idea seeing as you're keeping all your AC's out of the chassis which can only be good... even if not much better than alternative, any betterness has got to be better. just a question of how much money and time goes into any improvements we want to achieve, and how much they do improve our design... that is something important to think about. The DC heater suply is a good idea though, especially with a SE amp, and if you are going to be amplifying low level signals in the preamp, such as a record player... as phono inputs are very low, and any leakage from filaments, or even filament suply leads is bad, and will be very noticeable.

also, you said about improving high freequency response. what i'm going to say realtes to acutual construction, not design... but if you want to further improve high frequency resposnse even more, make sure all components and leads, particularly caps, in audio path are mounted well away from chassis. If not, you get a capacitance between the component/lead and the chassis.. as the chassis is earthed, this capacitance is to ground. any capacitance to ground is bad for HF response!

feedback will alter sound... as far as 'kill' the sound, i will also agree with your father. i do believe feedback can be good, but it can also be bad. it is something that must be designed around in order to be used for good.

For the best acuracy in sound reproduction, feedback in any stage is bad!!! this is true, as the feedback flattens the frequency response... this is probably what is meant by manufacturers who say that it kills the sound, as feedback will reduce the amount of detail in the sound... but, this can also a good thing depending on what you want. this flattening will lead to an extended frequency response, as it flattens, or at least makes smaller any 'humps'. so you will get extended low and high end.

Feedback also reduces the distortion of the stages within the feedback loop, and will give a more hi-fi sound because of this. The negative of this is the transition into distortion is abrupt, which will mean large peaks in the signal will be more noticeably clipped. this is because the feedback loop will keep the distortion to a minimum, until the tube is driven into clipping, and at this point, there is no excess gain to keep loop operating properly, the loop is therefore effectively broken, and we get a very abrupt transitin into distoriton.

Feedback in the output stage also has a big effect on the sound of the amp. The feedback reduces the impedance of the stage, and in an output stage, makes it less reactive to the reactive load of your speakers. This will make the amp sound tighter, and will also extend frequency response, but wether it is good or bad is subjective, as is the whole idea of feedback being bad... i am not going to tell you what is good and bad... it is all about listening, and if it sounds good, go with it. if it sounds bad, then don't! simple as that.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
813 final design? Richard Ellis Tubes / Valves 17 1st October 2008 03:58 AM
UcD Amp Design Paper arif__anwar Class D 2 20th March 2008 10:15 PM
KT88 pp triode control grid resistor???? sbcellis Tubes / Valves 3 24th February 2007 04:00 PM
Final Design inrank Full Range 8 1st September 2006 05:07 PM
Decoupling Paper In Oil capacitor question for KT88 design. Layberinthius Tubes / Valves 18 23rd March 2004 01:41 PM


New To Site? Need Help?

All times are GMT. The time now is 04:02 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2