New OTL circuit

[Joe Lau]

New circuit of OTL amplifier , Easy to DIY - pattern in China

This structure can be apply on difference tubes like 6H30 , 2A3 , 300B , 6H6 ...... or any load impedance tube. Use 6H30 , 2A3,300B can drive 32 Ohm Headphone to very high level & good sound. This circuit is for DIY only , not for commercial. It had pattern in China by me.

Dew

 

[Negatron]

Quote:

Originally Posted by Joe Lau /img/forum/go_quote.gif New circuit of OTL amplifier , Easy to DIY - pattern in China This structure can be apply on difference tubes like 6H30 , 2A3 , 300B , 6H6 ...... or any load impedance tube. Use 6H30 , 2A3,300B can drive 32 Ohm Headphone to very high level & good sound. This circuit is for DIY only , not for commercial. It had pattern in China by me. Dew

Hello, Good to see you back posting. Very interested in your contribution.
Those are usually very nice sounding circuits. Thanks.
Welcome back!

 

[EvilJan]

Looks very simple. I think it would be a good start into tubes, am I right?
But I found some minor failures in the Picture.. The parts list says that VB1 is a 12AX7, and right after that comes VB1,2,3 6DJ8. Should'nt it be only VB2 & 3? Also i'm a bit confused, did you want to save the place normally used by the heating?

Maybe I will try building this Amp. Is it also possible to replace VB2 & 3 by 2 ECLL800? I've got quite a few of these and ECC83 (12AX7)

 

[Negatron]

Quote:

Originally Posted by EvilJan /img/forum/go_quote.gif Looks very simple. I think it would be a good start into tubes, am I right? But I found some minor failures in the Picture.. The parts list says that VB1 is a 12AX7, and right after that comes VB1,2,3 6DJ8. Should'nt it be only VB2 & 3? Also i'm a bit confused, did you want to save the place normally used by the heating? Maybe I will try building this Amp. Is it also possible to replace VB2 & 3 by 2 ECLL800? I've got quite a few of these and ECC83 (12AX7)

You might want to price the ECLL800 before you use it in a DIY project, they may sell for enough to buy quite a few suitable tubes. Most any triode or triode wired pentode with a high gm and low rp will work fine as outputs. The $2 Russian 6H6P is a great tube for that with the spec'd R. The higher output impedance of that circuit does some interesting things with output voltage into lower impedance 'phones to self correct non linearity. Works better than it looks on paper with a smaller than usual OP cap.

 

[headphonejunkie]

Are you saying you would not need feedback as in th CJ amp for grados. That is interesting if true.

 

[Joe Lau]

Quote:

Originally Posted by EvilJan /img/forum/go_quote.gif Looks very simple. I think it would be a good start into tubes, am I right? But I found some minor failures in the Picture.. The parts list says that VB1 is a 12AX7, and right after that comes VB1,2,3 6DJ8. Should'nt it be only VB2 & 3? Also i'm a bit confused, did you want to save the place normally used by the heating? Maybe I will try building this Amp. Is it also possible to replace VB2 & 3 by 2 ECLL800? I've got quite a few of these and ECC83 (12AX7)

Send me detail data of ECLL800. I can check it possibility to use on it or not. It has some mistake on circuit. Original design is 6DJ8 x 3 but it seem to be not enough gain & modify to 12AX7 on input stage. I will send correct circuit tomorrow.

 

[Joe Lau]

Quote:

Originally Posted by Negatron /img/forum/go_quote.gif You might want to price the ECLL800 before you use it in a DIY project, they may sell for enough to buy quite a few suitable tubes. Most any triode or triode wired pentode with a high gm and low rp will work fine as outputs. The $2 Russian 6H6P is a great tube for that with the spec'd R. The higher output impedance of that circuit does some interesting things with output voltage into lower impedance 'phones to self correct non linearity. Works better than it looks on paper with a smaller than usual OP cap.

Yes ! You are correct. 6H6 is very good tube for this application. It has high power & low impedance. We will post a 6N3 + 6N6 (6H6) version diagram later. This circuit is very good for Headphone , Line amplifiers & buffer on CD. Gain can control by additional feed back.

 

[Joe Lau]

Quote:

Originally Posted by EvilJan /img/forum/go_quote.gif Looks very simple. I think it would be a good start into tubes, am I right? But I found some minor failures in the Picture.. The parts list says that VB1 is a 12AX7, and right after that comes VB1,2,3 6DJ8. Should'nt it be only VB2 & 3? Also i'm a bit confused, did you want to save the place normally used by the heating? Maybe I will try building this Amp. Is it also possible to replace VB2 & 3 by 2 ECLL800? I've got quite a few of these and ECC83 (12AX7)

We had build 6N3 (5670) + 6N6. It can drive 32 Ohm Headphone very well. It even better than 6DJ8. You also can use 6H30 , 2A3 , 300B or any other low impedance tubes. We are build 3 model base on 6DJ8 , 6N6 & 2A3 & show in Hong Kong Head-fi meeting & like by most visitors. 6H6 Version circuit is at bottom of 6DJ8 version now.

 

[Negatron]

Quote:

Originally Posted by headphonejunkie /img/forum/go_quote.gif Are you saying you would not need feedback as in th CJ amp for grados. That is interesting if true.

The simple answer is NO, you don't need feedback. But there IS feedback, but it is local, not global.
The degeneration through the un-bypassed cathode resistor is feedback.
Global feedback in a triode is considered by some, myself included, to be a band-aid. Global feedback lowers output impedance but adds other ill behaved temper to the circuit. This just scrapes the surface:
global feedback transient distortion - Google Search

Joes second circuit is a quasi SRPP but is in effect a cathode follower sitting on a very large cathode resistor. Should drive Grado's with no problem what so ever.

 

[amphead]

Welcome Back Joe Lau!
Looks like a good circuit! No beginners however, due to the high voltage. It takes some practice to safely build HV amps. The tubes are fixed bias, so one less step to complete the circuit. Edit: Are you doing a pcb version or point to point wiring? thanks

 

[EvilJan]

http://frank.pocnet.net/sheets/062/e/ECLL800.pdf

I'm sorry that the datasheet is written in german, but I think you will understand what is written there.

If you have any problems understanding this, ask me, I'm german too

some websites say that this Tube contains 2x EL95

Thank you for reviewing it!

 

[headphonejunkie]

So, this circuit could be better than a CJ amp.That is something to think about. I may just have to build this to hear how it sounds.This circuit is very similar to the morgan jones amp except the input goes into the output stage differently. I wonder how much of a difference in sound this makes.

Quote:

Originally Posted by Negatron /img/forum/go_quote.gif The simple answer is NO, you don't need feedback. But there IS feedback, but it is local, not global. The degeneration through the un-bypassed cathode resistor is feedback. Global feedback in a triode is considered by some, myself included, to be a band-aid. Global feedback lowers output impedance but adds other ill behaved temper to the circuit. This just scrapes the surface: global feedback transient distortion - Google Search Joes second circuit is a quasi SRPP but is in effect a cathode follower sitting on a very large cathode resistor. Should drive Grado's with no problem what so ever.

 

[Joe Lau]

Quote:

Originally Posted by amphead /img/forum/go_quote.gif Welcome Back Joe Lau! Looks like a good circuit! No beginners however, due to the high voltage. It takes some practice to safely build HV amps. The tubes are fixed bias, so one less step to complete the circuit. Edit: Are you doing a pcb version or point to point wiring? thanks

12AX7 + 6DJ8 is PCB version. 2A3 is Telfon Plate + connecting post in prototype. Will consider release 6N3 + 6N6 version in Kit by other brand name in China & Hong Kong.

 

[nikongod]

increase BOTH size and voltage rating of c14&15. in normal operation, the voltage is good, but if the top triode fails, the cap can see b+. 100uf is way small for a 32 ohm load - 470/350V caps are not tooo expensive.

no cathode resistor bypass caps?

I personally would also use R17/c11 (in the general schematic) in the applied one. the routing of the power supply is different. it should improve channel separation a decent bit.

 

[Joe Lau]

Quote:

Originally Posted by nikongod /img/forum/go_quote.gif increase BOTH size and voltage rating of c14&15. in normal operation, the voltage is good, but if the top triode fails, the cap can see b+. 100uf is way small for a 32 ohm load - 470/350V caps are not tooo expensive. - You are right. We will use 250V in production no cathode resistor bypass caps? I personally would also use R17/c11 (in the general schematic) in the applied one. the routing of the power supply is different. it should improve channel separation a decent bit.

Yes ! This circuit cathode resistor can not by pass. It is for local feedback. 100u -3db is around 25Hz. Sure 470u has better low frequency response. -3db will down to 5hz.