Jump to content

DNA (Donald North Audio) Headphone Amp


blubliss
 Share

Recommended Posts

I have talked with Peter at least once about a headphone amp. At one point, there was a plan for AN kits to release one, hes said, but for some reason it hasn't gotten off the ground yet. A silver-nickel output transformer for a headamp would be quite an undertaking at

Link to comment
Share on other sites

  • Replies 159
  • Created
  • Last Reply

Top Posters In This Topic

Great thread full of techno-babble (after more than 4 years into this hobby, I'm still a newb's newb :palm:) and humor (Smurfs FTW)! At least I learned the reasons for the low-z and high-z jacks on my Meier Corda amp. When I did briefly try the high-z with some of my headphones (I don't remember which ones), I didn't quite like the sound for the reasons recstar mentioned and never used it. But I'm going to experiment a little with my DT990 Manufaktur (600 ohm). Fun stuff!

... What I notice immediately is that the sound out of the high-z sounds more spacious and clear and a little more detailed with better transients. With the low-z output, the vocal was much more upfront by comparison, and the high-z sounds more balanced and natural.

I have one question. Is the gain switch on some amps an attempt to address the low/high-Z headphone factor?

Link to comment
Share on other sites

Regarding gain switch, to which amps are you referring? Can you list a couple examples?

Donald, when it comes to gain switches I think Young (socrates63) is referring to amplifiers that either through the use of series resistance on the output or resistors within the circuit manipulate the perceived gain of the amplifier. With the first example there are obvious downsides and pitfalls but lots of solid state designs are able to change the gain stage on the fly without adversely affecting performance. In my experience this is pretty difficult to do with tube amps short of adding a simple voltage divider (or similar) to the input section.

For commercial examples have a look at Headamp and Headroom, both of whom offer gain switches for most of their products.

Link to comment
Share on other sites

Rockhopper M^3 and Channel Island Audio VHP-2 are two solid state amps I currently own which have a gain switch.

The most obvious effect of the gain switch was on the volume. I perceived some differences in the sonic character of the sound, but they weren't too noticeable (or maybe I just imaged a difference because I expected it :palm:... I'm not too sure). However, with my Corda amp, there definitely were noticeable changes to the sound when using low-z- and high-z outputs.

Link to comment
Share on other sites

Donald, when it comes to gain switches I think Young (socrates63) is referring to amplifiers that either through the use of series resistance on the output or resistors within the circuit manipulate the perceived gain of the amplifier. With the first example there are obvious downsides and pitfalls but lots of solid state designs are able to change the gain stage on the fly without adversely affecting performance. In my experience this is pretty difficult to do with tube amps short of adding a simple voltage divider (or similar) to the input section.

For commercial examples have a look at Headamp and Headroom, both of whom offer gain switches for most of their products.

Or, if you have a really good attenuator for the volume, you can attenuate without too disastrous consequences. I have a "big" shallco ladder attenuator as the volume control on a tube preamp with a high-low attenuation switch and I hear hardly any sound degradation.

Link to comment
Share on other sites

  • 2 weeks later...
There are a bunch of things that could be done with this circuit to improve it for only marginally more money.
Such as?

Would these be marked sonic improvements, or just more complexity to the circuit?

I've built that amp (down to the JJ cap, the cerafine bypasses, the Edcor OPTs and the 5AR4 -- I think I used 6N6p's instead of the 6H30 ... basically the same tube) and it is fine, but there is plenty of room for easy improvement.
What would you do? How would it be better?
Link to comment
Share on other sites

Such as?

Would these be marked sonic improvements, or just more complexity to the circuit?What would you do? How would it be better?

I don't believe in complexity, other than where it is necessary. So, no, these changes would not greatly increase complexity. In fact, one of them would reduce cost by a couple of dollars.

To be sure we are on the same page, my guess as to the schematic is attached. Donald can chime in if I am wrong about something (oh, and there should be a resistor in series with the output I believe -- he said that a few posts back.)

The two circles represent the simplified signal current path. The primary sin here is that the input and output signals both pass through electrolytic capacitors. My experience is that the soft "tubey" sound of many tube amps is, in large part, actually the sound of the capacitors which impart their sound on the amp more than any other part. At the very least these should be film caps. But, even better is finding other solutions to the issue.

A first solution is to use some other method of bias. This can be an LED, a battery, or any number of options (see bias options here). Which ever the choice, almost all of them are better than the old resistor + cap.

The second issue is the power supply. Aside from the big cap in the signal path, this is a high Z supply that will sag with signal. Sag is great in a guitar amp, but not so great for hifi. A simple linear regulator will both quiet the supply and eliminate the sag. This sort of supply can literally be as simple as a single chip and a few resistors. The linear regulator still has a big cap on the output, but there are ways around that, too.

Indeed, one solution is a shunt regulator. Using a CCS load for it, you can have a low Z low noise supply, and this can eliminate the cap from the signal path as well as the CCS presents a high Z barrier. I did something like that here with a glow tube. Note that mine also obtains fixed bias with a resistor by running all the current, but none of the signal, though the biasing resistor.

The third issue is that I just don't think Edcor OPTs are all that great. I think they are fine for bread boards, but for a production amp, I'd expect something better. Something better costs more, and maybe that is the issue here, but if this is a budget amp then market it as such, not as an amp "with unprecedented fidelity and emotional involvement".

Finally, for low Z headphones, better bass is obtained with a better damping factor. The earlier stated parameters for this amp either require that the load on the tube is too low, or that the Z out is too high.

post-1078-1295115402424_thumb.jpg

Link to comment
Share on other sites

My choice of bias is based on sound quality. Yes, you can bias the tube with a battery or LED but it has not been my experience that these necessarily sound better, having their own non-linearities.

Film caps sound different be also not necessarily better. The caps I use are selected as part to achieve the overall sonic synergy.

As mentioned before, the output transformers are custom made to my specification. In my experience, Edcor makes very good sounding transformers independent of price. Transformer design is somewhat a "black art". I have heard transformers by some other well-regarded manufacturers which didn't catch my ear and sounded boring and uninvolving to me. Higher price does not guarantee better sound. I believe the Sonett is well balanced and value engineered to deliver the best sound possible for the price. In my opinion the next level up in output transformers are Audio Notes. To use their M4 EI core transformers will double the price of the amplifier.

Regarding power supply sag and constant current sources, do remember a good output transformer does double duty approximating a CCS. Also note the high voltages are decoupled between the 2 channels.

As I have mentioned before, in my design the tube is always presented with at least 3x the plate impedance as a load, even with low impedance headphones on the IEC setting.

Desired damping factor is a function of the headphone design. Some are designed for low impedance sources; others are meant to see a higher impedance source. In the Low impedance setting the output impedance is 28 ohms, yielding a damping factor greater than 1 with Grados. As has been mentioned in other threads, some listeners find little to no sonic benefit with >1DF on headphones designed for low impedance drive. Of course to each his own and listen for yourself.

Link to comment
Share on other sites

Anyone else besides me wonder who oatmeal769 is?

USG

Most of us, probably. I suspect that he got much more of an answer than he expected. Now he either has to be quiet or be very technically savvy in a response. I predict that the former will be the case, but I've been known to be wrong from time to time. :P

Link to comment
Share on other sites

As I have mentioned before, in my design the tube is always presented with at least 3x the plate impedance as a load, even with low impedance headphones on the IEC setting.

The 6H30 has a gain of around 15 and a plate impedance of 1500 ohms. So, let's run the math.

Let's pick a Grado with a 32 ohm impedance, for this to present a load that's 3x the Rp of the tube, you'll need to get the impedance up to 4500 ohms, an impedance ratio of 140:1 which requires a turns ratio of 11.9:1 on the transformer.

This would give an overall voltage gain of 2dB, which is a fair bit short of the 6dB of gain which is claimed for the IEC output.

Which means 1 of 3 possibilities:

  1. The transformer doesn't present an impedance of 3x the tube's plate impedance
  2. The gain is 2dB, not 6dB
  3. Your idea of a low impedance headphone is something with an impedance of around 80 ohms
Link to comment
Share on other sites

With my Amplitrex, for the 6H30 I measure a plate impedance around 1300 ohm at my operating point.

For the IEC output, you are forgetting that I add a series resistor on the output to bring the output impedance up to 120 ohm. And yes, my minimum 3x plate impedance claim is based around 32 ohm Grados. However I suspect most people with Grados will probably use the Low setting with 28 ohm output impedance. In this situation, the plate load impedance is around 4.5x.

On the IEC setting with 600 ohm headphones I do measure gain about 6dB. With low impedance headphones, the gain of course will be much less.

Link to comment
Share on other sites

With my Amplitrex, for the 6H30 I measure a plate impedance around 1300 ohm at my operating point.

For the IEC output, you are forgetting that I add a series resistor on the output to bring the output impedance up to 120 ohm. And yes, my minimum 3x plate impedance claim is based around 32 ohm Grados. However I suspect most people with Grados will probably use the Low setting with 28 ohm output impedance. In this situation, the plate load impedance is around 4.5x.

On the IEC setting with 600 ohm headphones I do measure gain about 6dB. With low impedance headphones, the gain of course will be much less.

Just wondering?

Is oatmeal769 a friend of yours like mike1127 is?

USG

Link to comment
Share on other sites

By the way, while I will say that I disagree with just about every assertion that Donald makes in his post above (regarding bias; caps; OPT specs, manufacturers, and core material; damping; whether OPT design is a black art; optimal tube load; etc.), my interest is not in derailing Donald's sales. As noted, I do things very differently. However, this doesn't mean I think what he does is wrong, and we still probably agree more than we disagree. I like transformer coupled amps, and I am glad to see that they are becoming available. Even with my complaints, it is still topologically a step up from the SP and RSA amps of the world (and it's easy enough to mod ;) ). So, I said my peace, and I will keep further discussions of this issue offline.

Link to comment
Share on other sites

By the way, while I will say that I disagree with just about every assertion that Donald makes in his post above (regarding bias; caps; OPT specs, manufacturers, and core material; damping; whether OPT design is a black art; optimal tube load; etc.), my interest is not in derailing Donald's sales. As noted, I do things very differently. However, this doesn't mean I think what he does is wrong, and we still probably agree more than we disagree. I like transformer coupled amps, and I am glad to see that they are becoming available. Even with my complaints, it is still topologically a step up from the SP and RSA amps of the world (and it's easy enough to mod ;) ). So, I said my peace, and I will keep further discussions of this issue offline.

That's too bad. I'm sorry you feel that way. :sadcat:

Maybe you'll change your mind? ;D

USG

Link to comment
Share on other sites

Did I miss a revision or something, cause it looks even more tacky and stupid than I remember.

This is the non-burn up version I believe. :D It's incredibly tacky to say the least...

That was the special trose49 amp, the one he was bragging about right before he ripped off some poor kid for a lot of money and disappeared from the forum.

Link to comment
Share on other sites

Anyone else besides me wonder who oatmeal769 is?... " Is oatmeal769 a friend of yours like mike1127 is?
I am not. I do however intend to make Donald's acquaintance and listen to his amp this Saturday at a meet in Northern LA. I am in the market for a step up from what I have now. I mentioned somewhere that I'd heard the EC Zana Deaux at Can Jam and that it had been my 'favorite' so far. Don PM'd me asking me to try his. I'm here doing research on my upcoming purchase, and his amp is in the running.
My naive guess would be he's the same oatmeal769 on the other site.
That is correct. I have found more dis-information there than good. I'm hopeful searching here may shed better light. I wouldn't post at all, but I've been following this thread keenly.
... I suspect that he got much more of an answer than he expected. Now he either has to be quiet or be very technically savvy in a response. I predict that the former will be the case
The former is indeed the case. I know very little about the guts of an amplifier. What I do know is from years of listening rather than electrical theory. I know what sounds good to me, and more importantly what doesn't.

I know that I agree with what's said about caps coloring as much or more than anything else, because I've swapped them on my own stuff and heard it. I know that I'm one who likes more damping factor than less with bass, and I'm pretty big on a quick slew rate, and enough power reserve to render accurate transient response and dynamics. I only know this from what I've listened to, and been shown by others with far more knowledge than me. I know I like the quietest noise floor possible, but I've no idea how that's done. I've also heard power lag or sag in tube amps, because I am also a formerly pro musician, and sound guy. Good for blues guitar, bad for headphones.

I also know that the sound of a setup is the sum of it's parts. An amp must be chosen based on the speakers with which it will work. I'm mostly concerned with how this amp will work with my HD600's, (and maybe a set of Grado's down the road) which I like best in the 'under a grand' club.

Can I follow the nitty-gritty of what's being spoken about electrically in this thread? Heck no. But I do know that it all relates directly to the things I've just mentioned. I hope you all won't mind if I lurk a bit longer. I'd also be REALLY interested in listening to an amp closer in design to the alternatives spoken of here, if someone might point one out.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share


×
×
  • Create New...