Author Topic: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier  (Read 48748 times)

0 Members and 1 Guest are viewing this topic.

Offline helius

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: us
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #25 on: January 07, 2016, 03:23:51 am »
So, in our world of programmable analog circuit arrays and perfect rail-to-rail opamps, when would you *need* to use a tube?  When does using silicon not make sense?
http://spectrum.ieee.org/semiconductors/devices/the-quest-for-the-ultimate-vacuum-tube
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1117
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #26 on: January 07, 2016, 03:58:43 am »
I saw it was a pentode and though "WTf I though you used single end triodes for audiophile cred, pentodes and push-pulls got the distortion too low" 

Then dave had the schematic drawn out...   :-DD
 

Offline boffin

  • Supporter
  • ****
  • Posts: 861
  • Country: ca
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #27 on: January 07, 2016, 04:18:54 am »
Could you just pull the valve and connect the grid to the anode?

 

Offline Pentium100

  • Regular Contributor
  • *
  • Posts: 245
  • Country: lt
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #28 on: January 07, 2016, 04:20:05 am »
I saw it was a pentode and though "WTf I though you used single end triodes for audiophile cred, pentodes and push-pulls got the distortion too low" 

Then dave had the schematic drawn out...   :-DD

Pentode in SE usually has higher distortion. Connecting it in ultra-linear mode lowers the output power and distortion, triode mode is lowest power and lowest distortion.

A JFET has IV curves almost exactly like those of a pentode, however, I do not know of any semiconductor device that has the IV curves of a triode.
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 9215
  • Country: au
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #29 on: January 07, 2016, 04:23:28 am »
Could you just pull the valve and connect the grid to the anode?

Not in this design.  The anode goes to the positive rail.

You'd have to connect the grid to the cathode - but, yes, your basic idea is ok.

Except for one thing - the signal would no longer go through a valve, removing the premise for it being there.
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 9215
  • Country: au
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #30 on: January 07, 2016, 04:28:55 am »
The fact that the valve doesn't effectively do anything isn't 'relevant' to the target market.  The fact that the signal does, in fact, pass through a valve is enough for them.


Also, I find the ultra low distortion figure absolutely hilarious when that target market is after the valve 'sound'.
 

Offline boffin

  • Supporter
  • ****
  • Posts: 861
  • Country: ca
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #31 on: January 07, 2016, 05:06:23 am »
Could you just pull the valve and connect the grid to the anode?

Not in this design.  The anode goes to the positive rail.

You'd have to connect the grid to the cathode - but, yes, your basic idea is ok.

Except for one thing - the signal would no longer go through a valve, removing the premise for it being there.
Yes, I meant Cathode; damn electron flow...
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1117
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #32 on: January 07, 2016, 05:52:51 am »
I saw it was a pentode and though "WTf I though you used single end triodes for audiophile cred, pentodes and push-pulls got the distortion too low" 

Then dave had the schematic drawn out...   :-DD

Pentode in SE usually has higher distortion. Connecting it in ultra-linear mode lowers the output power and distortion, triode mode is lowest power and lowest distortion.

A JFET has IV curves almost exactly like those of a pentode, however, I do not know of any semiconductor device that has the IV curves of a triode.
Yeah I'm not up on valve amps, I thought I had read somewhere that as they progressed the distortion got down quite low (I forget how, thought it was pentodes, apparently not. ) but in doing so, lost the tube amp 'sound'
 

Offline blueskull

  • Supporter
  • ****
  • Posts: 12302
  • Country: cn
  • Power Electronics Guy
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #33 on: January 07, 2016, 05:58:54 am »
yay for Chinese numbering.  Same "number", but completely different part thats not even remotely close to the original.  Nothing like swapping a triple triode for a sharp cutoff pentode and sticking the same number on it for confusion reasons.

Honestly the reason that solid state makes no sense to me is because I get electron flow.  "conventional" flow is backwards and makes no sense in my head.

All Chinese tubes made before 80s are direct copy of Soviet tubes, hence having soviet part numbers.

They are named in xyz form, where x and z are numbers, same to the numbers in the Soviet version, where y is an English letter, that corresponds to a Russian letter.
 

Offline blueskull

  • Supporter
  • ****
  • Posts: 12302
  • Country: cn
  • Power Electronics Guy
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #34 on: January 07, 2016, 06:04:47 am »
Tubes can affect audio characteristics a lot, pretty much by introducing harmonics and compress input level.

Some people found listening to slightly distorted music feels more comfortable, hence the popularity of tube amp.

Personally I prefer exact sound reproduction, not distorted, hence I use high end DAC coupled with monitor grade IEM.

But I know a few people prefer tube or JFET amplified sound that introduces harmonics. They claim it sounds warmer.

There are even commercial VST plugins to digitally introduce tube effects to a piece of sound track.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 4607
  • Country: gb
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #35 on: January 07, 2016, 06:15:36 am »
Personally I prefer exact sound reproduction, not distorted, hence I use high end DAC coupled with monitor grade IEM.
What do you consider exact sound reproduction? Our hearing is tied in with our vision, so without the original images we don't hear things as we would in real life. The McGurk effect is the most famous example of this, but it isn't just a speech issue. It happens with musical instruments and other sound sources too. Also, those inner ear monitors might reproduce the super low bass from the lowest pedals of a cathedral organ, but they won't give you that punch in the chest you feel in the actual cathedral, so that's not faithful reproduction, either.
« Last Edit: January 07, 2016, 06:28:29 am by coppice »
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 9215
  • Country: au
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #36 on: January 07, 2016, 06:26:06 am »
Could you just pull the valve and connect the grid to the anode?

Not in this design.  The anode goes to the positive rail.

You'd have to connect the grid to the cathode - but, yes, your basic idea is ok.

Except for one thing - the signal would no longer go through a valve, removing the premise for it being there.
Yes, I meant Cathode; damn electron flow...

Nice try, but I'll call you on that.   :D
 

Offline blueskull

  • Supporter
  • ****
  • Posts: 12302
  • Country: cn
  • Power Electronics Guy
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #37 on: January 07, 2016, 06:26:38 am »
Personally I prefer exact sound reproduction, not distorted, hence I use high end DAC coupled with monitor grade IEM.
What do you consider exact sound reproduction? Our hearing is tied in with our vision, so without the original images we don't hear things as we would in real life. The McGurk effect is the most famous example of this, but it isn't just a speech issue. It happens with musical instruments and other sound sources too. Also, those inner ear monitors might reproduce the super low bass from the lowest pedals of a cathedral organ, but they won't give you that punch in the chest you feel in the actual cathedral, so that's notfaithful reproduction, either.

My standard:

DAC and HPA does not introduce any flavor, headphone has flat response compared to spl.
 

Offline mikerj

  • Super Contributor
  • ***
  • Posts: 2095
  • Country: gb
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #38 on: January 07, 2016, 09:30:21 am »
A big fail is that any ripple on the -12v switcher rail is inejcted into the input of the op-amp stages, something you can clearly see on the scope when the amp is powered up.  I was hoping Dave was going to put the scope into FFT mode jut to see what kind of crap was coming out of the valve compared to the output of the sig gen.
 

Offline Salas

  • Frequent Contributor
  • **
  • Posts: 291
  • Country: gr
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #39 on: January 07, 2016, 09:45:14 am »
How high is the anode voltage used and how is it created in this headphones amp?
 

Offline kassu

  • Newbie
  • Posts: 1
  • Country: fi
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #40 on: January 07, 2016, 10:06:29 am »
I think we are all missing the point here. This was sold for about 50 bucks. No true "audiophile" would buy a 50 dollar headphone amp, so this is not "audiophile quality".

Now, if you would sell the same thing for 500 dollars, it would be a different story...
 

Offline mikron

  • Contributor
  • Posts: 38
  • Country: de
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #41 on: January 07, 2016, 10:21:40 am »
I wonder how much these cheap ceramic caps in the signal path will improve the listening experience...
 

Online daqq

  • Super Contributor
  • ***
  • Posts: 1553
  • Country: sk
    • My site
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #42 on: January 07, 2016, 10:38:02 am »
Dave, didn't you have some dynamic spectral analyzer? It might be interesting to look with it at what comes in vs what comes out.
Believe it or not, pointy haired people do exist!
+++Divide By Cucumber Error. Please Reinstall Universe And Reboot +++
 

Offline G7PSK

  • Super Contributor
  • ***
  • Posts: 3642
  • Country: gb
  • It is hot until proved not.
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #43 on: January 07, 2016, 11:12:45 am »
Funny thing is that valve is not in my 1969 Mullard data book, but it is the 1959 book under miscellaneous but there is no data.
 

Offline mikerj

  • Super Contributor
  • ***
  • Posts: 2095
  • Country: gb
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #44 on: January 07, 2016, 11:14:04 am »
How high is the anode voltage used and how is it created in this headphones amp?

About 24v total (-12 on cathode, +12 on anode) and two small switching regulators are used to create the voltage rails.
 

Online ciccio

  • Frequent Contributor
  • **
  • Posts: 619
  • Country: it
  • Designing analog audio since 1977
    • Oberon Electrophysics
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #45 on: January 07, 2016, 11:43:44 am »
I don't want to discuss the electronic design or the performance of this toy.
It obviously was designed for a market objective, and the design is done reasonably well.
What i discuss is what none of the above posters noted:
this toy (and a lot of other tube audio gear on the market) is designed and built without any consideration about product safety.
EN60065 (safety requirement for audio, video and similar electronic apparatus) and it's equivalent regulations are violated in at least 2 points:
- There are HOT parts accessible to the user (the tubes)
- There is a mechanical danger of breaking the glass tubes and cut yourself with the glass.

This toy should not be sold, even in China.

Best regards
Ciccio

Strenua Nos Exercet Inertia
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2528
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #46 on: January 07, 2016, 12:08:31 pm »

- There are HOT parts accessible to the user (the tubes)
- There is a mechanical danger of breaking the glass tubes and cut yourself with the glass.

This toy should not be sold, even in China.

Best regards

Well, you guys heard ciccio, mains voltage and glass objects are dangerous!

We better get rid of wall outlets, since the HOT parts are accessible to users... We should also do away with glass cups and bottles as we could cut ourselves!

In fact, personal Faraday cages and sippy cups should be mandatory for everyone to prevent electrocution or severe gashes!
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline dr.diesel

  • Super Contributor
  • ***
  • Posts: 2123
  • Country: us
  • Cramming the magic smoke back in...
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #47 on: January 07, 2016, 12:17:49 pm »
EN60065 (safety requirement for audio, video and similar electronic apparatus) and it's equivalent regulations are violated in at least 2 points:
- There are HOT parts accessible to the user (the tubes)
- There is a mechanical danger of breaking the glass tubes and cut yourself with the glass.

This toy should not be sold, even in China.

 :palm:

Let's not dumb down society even more, complete fail.

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 9965
  • Country: 00
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #48 on: January 07, 2016, 12:39:26 pm »

I discovered a long time ago it makes no difference which way the little packets of energy go, it all works out just fine. :)

Also just too many parts to keep track of :) 

24v K-A and a cathode follower arrangement?  Whats the point?  It would give lower output impedance, but feeding into an op-amp its not like that would be required. Not enough voltage to make any useful gain anyway.  Wonder how far out of the linear operating range it is.
In my opinion there is no point, That is not a good tube for a cathode follower, the designer is not taking advantage of the high input impedance a vacuum tube offers, and as you pointed out the plate voltage (B+) is too low.

The designer isn't even trying to prove something.

Why not?
build a 30W headphone amplifier using all tubes, no output transformer, using an output differential amplifier with cathode output. (to keep the voltage and impedance low)? Run it on say............125V regulated(+/- 65.5V).

That would be a concept amplifier worth building, if for no other reason than to prove the idea. :)

 Why 30 watts, not many headphones can accept even peaks at that level? About the only thing that might differentiate a 'headphone amp' form other audio amps is that it might be designed to only work at the higher output impedance of most headsets.

Not many headphones can accept 30W RMS, but 30W peaks? Sure.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 9965
  • Country: 00
Re: EEVblog #837 - Reverse Engineering A Valve Headphone Amplifier
« Reply #49 on: January 07, 2016, 12:41:07 pm »
No frequency response/distortion/noise measurements?  Am leaving disappointed.

And why would you "buffer" something that's going to be fed into an op-amp? The op-amp impedance isn't high enough?  :-//

Still, I guess I'm missing the point. I'm not the target market for this. I'm sure looking at glowing tubes will improve the sound quality for many people.


« Last Edit: January 07, 2016, 12:45:43 pm by Fungus »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf