Author Topic: Mars 5-10 stereo - update  (Read 18038 times)

Offline handsome

Re: Mars 5-10 stereo - update
« Reply #105 on: October 20, 2020, 10:23:44 AM »
while you in there you should drop that series resistor on the first tube's grid to about 1k as well. If you worried about RF add a 220-470pf cap from said grid to ground.

Offline El Sid

Re: Mars 5-10 stereo - update
« Reply #106 on: October 20, 2020, 02:38:51 PM »
I'm certainly not disagreeing with handsome's suggestion of an ECC82, and i'm not going to provide you wuth the new component values  ;D,  but you may also want to consider an E88CC. It also has less gain than the ECC83 (u=33) but plenty enough in this application. I like it because it is super-linear, and can also easily ne used for DC coupling as it is quite happy with a low anode voltage. I used it as the pre/splitter combo in my EL84 concept amp and am very pleased with it.

No less a luminary than the esteemed Spiderman uses it in the VA Exclame in its 6922 guise, so it must have something going for it....

Offline Qualityten

Re: Mars 5-10 stereo - update
« Reply #107 on: October 22, 2020, 08:23:31 PM »
See the modification to the Leak Stereo 20 here: http://lilienthalengineering.com/100-amplifiers-chapter-1/100-amplifiers-part-4.  Just scroll down for the discussion.

Offline charles

Re: Mars 5-10 stereo - update
« Reply #108 on: October 22, 2020, 09:05:39 PM »
My advise. Leave the Leak Stereo 20 as it is.

It sounds very good and it is a good design.

The day will come that somebody wants to modify a Leak 12 point one and we will call it.........................

Offline Mars

  • Commercial Member
  • Trade Count: (+1)
  • *****
  • Posts: 1,309
  • Total likes: 116
  • “The path is the goal.” Mahatma Ghandi
    • Mars Tube Audio
Re: Mars 5-10 stereo - update
« Reply #109 on: August 17, 2021, 06:52:53 PM »
Hi Everyone

I am rebuilding a Mars 5-10 for a client that wants the gain to tamed. I'll be posting some photos below. But first some technical details.

I had some good results using a split load inverter to bring down the gain to a more use-able level. Initially I used the ECC82 with good results. Studying the matter I came upon these articles that helped me along the path of improving the EL84 amplifier.

https://www.angelfire.com/electronic/funwithtubes/more_phase_inverter.html

This article made me realise how good the ECC88 is as a split load inverter. It also discourage me to use the ECC88 for the voltage amplifier section.

http://www.valvewizard.co.uk/cathodyne.pdf

More of the same, but an example of using a single ECC88 for both voltage amplification and phase inverting (DC coupled)

For this build I will stay with the EF86 on input duties. Here is the schematic:






OEM manufacture of Tube Audio
Power and Integrated Amplifiers
Phono Pre-amplifiers
DIY Kits
DIY Workshops

Offline Mars

  • Commercial Member
  • Trade Count: (+1)
  • *****
  • Posts: 1,309
  • Total likes: 116
  • “The path is the goal.” Mahatma Ghandi
    • Mars Tube Audio
Re: Mars 5-10 stereo - update
« Reply #110 on: August 18, 2021, 12:26:47 PM »
This is how the kit wiring looks:



I decided to do a complete rebuild and stripped the amp:

OEM manufacture of Tube Audio
Power and Integrated Amplifiers
Phono Pre-amplifiers
DIY Kits
DIY Workshops

Offline Mars

  • Commercial Member
  • Trade Count: (+1)
  • *****
  • Posts: 1,309
  • Total likes: 116
  • “The path is the goal.” Mahatma Ghandi
    • Mars Tube Audio
Re: Mars 5-10 stereo - update
« Reply #111 on: August 26, 2021, 05:14:35 PM »
Here is the schematic as it was built over the weekend.

Some values were left out:

  • R27 and R28 should be 47 ohm 5 watt
  • R7 and C7 left out for the moment till square wave tests
  • R6 was chosen at 33k. (39k or 47k might be better)
[li]R11 was chosen as 4K7. (2K2 might be better)
[/li][/list]

OEM manufacture of Tube Audio
Power and Integrated Amplifiers
Phono Pre-amplifiers
DIY Kits
DIY Workshops

Offline Mars

  • Commercial Member
  • Trade Count: (+1)
  • *****
  • Posts: 1,309
  • Total likes: 116
  • “The path is the goal.” Mahatma Ghandi
    • Mars Tube Audio
Re: Mars 5-10 stereo - update
« Reply #112 on: August 26, 2021, 05:17:01 PM »
Here is progress on the rebuild I am currently busy with:



PI filter in place. Earth bar across the power amp section in place.
OEM manufacture of Tube Audio
Power and Integrated Amplifiers
Phono Pre-amplifiers
DIY Kits
DIY Workshops

Offline Mars

  • Commercial Member
  • Trade Count: (+1)
  • *****
  • Posts: 1,309
  • Total likes: 116
  • “The path is the goal.” Mahatma Ghandi
    • Mars Tube Audio
Re: Mars 5-10 stereo - update
« Reply #113 on: August 27, 2021, 04:06:36 PM »
Update on circuit:

R11 kept at 4K7 ohm
C3 enlarged to 270uF 400V (220uF should be good too)
R6 changed back to 68K (improves split load inverter load match)
C12 47pF

I took some voltages:

B+ 310V


+/- 12V across 390 ohm cathode resistors



6.3V filament voltage with 230V AC mains



282V on phase inverter



+/- 70V across phase inverter anode load resistor



+/- 70V across split load inverter cathode resistor



+/- 140V across phase inverter triode



156V on EF86



65V on EF86 anode






OEM manufacture of Tube Audio
Power and Integrated Amplifiers
Phono Pre-amplifiers
DIY Kits
DIY Workshops

Offline Ampdog

  • Trade Count: (+1)
  • AVForums Grandmaster
  • *****
  • Posts: 7,584
  • Total likes: 436
  • Straight from the horses mouth real truths
Re: Mars 5-10 stereo - update
« Reply #114 on: August 29, 2021, 08:40:40 AM »
With due respect to to Charles, the circuits of both the Leak 2020 and TL12 are such general circuits that I do not view the suggested changes  so much as changes to the particular circuits, rather than ideas pointing the way forward at the hand of said circuits for further designs. Many amplifiers  followed these topologies, the Mullard 510 and 520 to name just two.

That said, the shown 2020 schematic still indicates the EL84 grid resistors as 1M, an error for which Leak paid dearly. This value falls outside the maximum allowed grid return values of 470K, only quoted as such in later data sheets. (It was found that values exceeding that caused grid run-away conditions with certain brands, as a result of which dozens of 2020's had to be recalled. Here some valve manufacturers themselves were initially in error!).
Audio must be the only branch of engineering where lack of basics' knowledge is considered a superior form of wisdom. (Anon)

Offline fredeb

Re: Mars 5-10 stereo - update
« Reply #115 on: August 29, 2021, 07:40:00 PM »
Hi Karel . Good stuff !

Have you considered running the EF86 in triode mode ?

Are you using only 1 half of each 6N1P ? What about the DC coupled long-tailed pair perhaps ? http://www.valvewizard.co.uk/dcltp.html
















« Last Edit: August 29, 2021, 07:44:12 PM by fredeb »
...evolution is the gradual development and stratification of progressive series of wholes, stretching from the inorganic beginnings to the highest level of spiritual creation.
Jan Smuts

Offline Mars

  • Commercial Member
  • Trade Count: (+1)
  • *****
  • Posts: 1,309
  • Total likes: 116
  • “The path is the goal.” Mahatma Ghandi
    • Mars Tube Audio
Re: Mars 5-10 stereo - update
« Reply #116 on: September 05, 2021, 11:20:58 AM »
Hi Ari

Thanks for the comment. I am only using one half of the 6N1P. I still need a bit of gain to lower the output impedance with global negative feedback, therefor the high gain of EF86 pentode works fine.

Great discussion here: https://www.tube-town.net/cms/?Info/How_To_Use_a_EF86

I am going to try the triode mode soon.

This is the triode configuration I want to try:



But first I want to try out pentode mode on the EL84's with this particular set of OPT.
OEM manufacture of Tube Audio
Power and Integrated Amplifiers
Phono Pre-amplifiers
DIY Kits
DIY Workshops

Offline fredeb

Re: Mars 5-10 stereo - update
« Reply #117 on: September 06, 2021, 08:27:23 PM »
Hi Karel

That triode mode for EF86 you show above seems like an excellent idea . Such low distortion negates the need for NFB - good one !

...evolution is the gradual development and stratification of progressive series of wholes, stretching from the inorganic beginnings to the highest level of spiritual creation.
Jan Smuts

Offline Ampdog

  • Trade Count: (+1)
  • AVForums Grandmaster
  • *****
  • Posts: 7,584
  • Total likes: 436
  • Straight from the horses mouth real truths
Re: Mars 5-10 stereo - update
« Reply #118 on: September 20, 2021, 10:16:34 AM »
Hi Karel

That triode mode for EF86 you show above seems like an excellent idea . Such low distortion negates the need for NFB - good one !


Hi Fredeb!

Careful here! The distortion figure given is just for the input and driver part! The main distortion comes from the power valves and output transformer!

Then again I am not too sure why everybody wishes to go for as little NFB as possible. Yes: Too much NFB increases the possibility of instability, but other than that there is no disadvantage. one must remember the effect of NFB on damping factor and valve differences. The designer has absolute control over the effects of NFB. (I hope that ancient lore like 'feedback sounds bad' and such is not still entertained.)
 
 
Audio must be the only branch of engineering where lack of basics' knowledge is considered a superior form of wisdom. (Anon)

Offline Qualityten

Re: Mars 5-10 stereo - update
« Reply #119 on: September 21, 2021, 05:06:23 PM »
I have been thinking about optimal NFB in my build of Karel's earlier 5-10 design, using ECF82s combining first stage amplification and phase splitting duties.  There his schematic calls for a 100K NFB resistor, with no capacitor bypass.  As 100K is much more than I am used to seeing in this place, I wonder whether this means that there is little NFB being applied?

In my particular build I am using 8K Vortexion output transformers intended for an EL84 valve amplifier and I can see oscillation or ringing when viewed on a scope with a 10KHz square wave.  As shown in Figure 5 example (d) in: https://www.rfcafe.com/references/radio-news/practical-techniques-square-wave-testing-july-1957-radio-tv-news.htm.  Should I add a capacitor in parallel with the 100K to control this?

Thanks for this interesting thread.