Neurochrome HP-1

Well... Use a bigger potentiometer, so you get better centering and see how big you can go before it overhangs. Get TKD pots and add the price bump to the assembled amp final price. Or bump it some extra, because in that case the PCB only offer will seem more valuable.

 

fishski13,

Thanks for the tips. I'm gonna have to search for an automatic center punch...I like the sound of that - I need one !

I just bought a Dremel Moto Saw from Amazon with metal blades. Should arrive Friday. I will report whether or not it makes clean square and xlr holes.

For round holes, I measure with digital calipers, mark the center crosshair point, punch a 0.5mm guide "divot," and drill a 1/16 guide hole with a regular drill; a Dremel is too high speed and too low torque, burns up the bit. Then use progressively larger bits until I have a hole that accomdates a step bit, drill to the closest size, and fine tune with a rounded metal file.

That's DIY, Baby!

 

Excellent. What I really like about this description of the HP-1 is that it essentially means that you can really hear the differences between headphones in perhaps greater clarity than you could with a more colored, less neutral amp, assuming you have a sufficiently resolving DAC (and if you are buying an HP-1, why wouldn't you have an awesome DAC to go with it?).

I know folks prefer and place emphasis on certain types of listening experiences but for me I would, 99% of the time, choose the most neutral amp possible with the best resolving/most detail oriented DAC that I could afford knowing that any differences I hear from that point on are mainly down to the headphone choice.
In other words, start your "painting" with a blank background (neutral source gear) and allow the individual headphones to add the color(s) if any are desired. I find such a setup (in my mind at least) very appealing.

 

A TKD pot adds significant cost percentage-wise to the BOM, and from what I can see on paper doesn't enhance the amp's performance. Little amp. Little knob. It works.

 

While the TKD pots certainly are nice, the $40 price tag holds me back as well, in particular as it doesn't enhance the technical performance of the amp. Maybe in that brushed aluminum $3k version of the HP-1, a TKD pot will be appropriate. Not because it would enhance the technical performance measurably, but solely because of the bling factor (which is important in purchasing decisions).

I did look at the mechanical dimensions of the TKD pot. As far as I can figure, it would require the front panel hole to move up a bit, so the various components would align to an even lesser extent on the front panel. That is incongruent with some of the critique I received a few pages back.

An ALPS RK271 would put the XLR and the volume knob at the same level. Unfortunately, the RK271 is usually not stocked well in the 2-10 kΩ range needed for the HP-1. I am not willing to trade off technical performance (noise floor in that case) just to get a bigger volume knob.

I appreciate the feedback. I'll do my best to make most people happy most of the time, as I can't make everybody happy all the time.

Tom

 

Amusingly, at $3k people are gonna scoff that the $40 pot isn't bling enough and will want a $400 stepped option, or some sort of LDR, or relays, or crosscoupled variable gain input stage, yadda yadda.

 

Yep. Some want bling. You then give them bling and they proceed to complain that it's not blingy enough. I won't go down that rabbit hole.

I don't get the attraction of LDR volume controls. I'll post measurements of the Tortuga LDR3 after I post the measurements of the O2. It'll take me a bit to get the LDR3 writeup done as I have a lot to say. Not much of it positive.

Tom

 

Goodness gracious! Passing our audio signal through an unstable variable component, controlled itself by another unstable variable component, the power of which is controlled in turn by the pots that they were trying to avoid in the first place, all subject to all the wiles and voodoo of power supplies etc. I don't see how that could possibly have gone wrong!

 

Selenium dioxide is not the most linear resistive material that I can think of. It also turns out that the input capacitance of the LDRs is rather nonlinear, so the bandwidth of the volume control changes (within the audio band!) as you turn the volume knob. The end result is 1-10 % THD even at line level as you turn the volume pot. The Tortuga LDR3 doesn't meet even DIN45500 and that's a rather old "hifi" standard.

The O2 measurements & review are up on my website (http://www.neurochrome.com/o2-headphone-amp/). Those interested in discussion of these measurements please follow me to this thread: http://www.superbestaudiofriends.org/index.php?threads/o2-measurements-review.3793/

Tom

 

We've talked briefly about this at the headphone meet-up in Calgary, but I am a big proponent of stepped attenuators (especially ladder types), not that there's anything wrong with potentiometers as long as they are quiet and the channels are relatively well matched.
Having said that, I don't like big price tags and overly blingy components, so I build all my own attenuators from scratch, every time.
I like to use Vishay RN55 resistors for my fancy ladder attenuators and Yageo general metal films for the series attenuators I build for Natayo (parts cost on this one is about the same as an Alps RK27 potentiometer, which I would have used instead if it was easy to get in 10K ohms).

P.S. love the blue chassis, btw. Black gets so boring sometimes.

 

Like reading the reviews so far. Curious

What sort of rotary switches do you use?

 

Nothing too special. Just a non-shorting, 24 position switch with silver plated leads:

 

I'm glad you like the blue chassis. Thanks. Good to hear from you again also.

I generally find that I end up towards the low volume (high attenuation) end of the attenuator enough that I end up wanting a step in between what's there. That's what drove me away from attenuators years ago. They do offer, by far, the best channel matching, though.

I think if I was going to implement a volume control solution other than a pot, I'd use one of the volume control chips out there. CS has a pretty nice one. Ah... So many projects, so little time.

Tom

 

How about doing negative gain? Frees the volume control for easier to drive stuff.

 

I'm not sure what you mean by "negative gain". Inverting? Attenuating? I'm also not sure what you mean about freeing the volume control.

An attenuator (be it with switches, pots, or a resistor ladder on an IC) will always degrade the SNR as it reduces the signal while adding noise. Physics is inconvenient that way. I do my best to minimize the degradation such that the noise added by the volume pot is comparable or below what the active circuits add.

The only way I can see improving the noise floor of the HP-1, which is 1.0 µV (20 Hz - 20 kHz integration bandwidth) already, would be to go with 1x gain rather than 2x gain in the output stage. I looked at that with various architectures and could not make it stable. Amplifier stability trumps noise floor; in particular when the noise floor is already super low.

Tom

 

C'mon Tom, obviously we need a cross coupled balanced output config with quad mono power supply (one each for the input and output stages) with upgradeable attenuator options. When are you gonna make us your "statement" amp?

 

Yeah I meant attenuating. By freeing I meant giving a larger usable range of volume control.

 

Ah. OK. You probably found the answer to your question reading my comment on the SNR of attenuators then.

Coming right up!

I suspect the HP-1 will be my statement headphone amp for a while to come. That and the DG300B for those who build it as a headphone amp. Two very different amps. Both statement amps - but different statements...

Tom

 

Tom,

I have a friend interested in HP-1 but he has minimal experience working with SMDs. He bought a practice kit from Amazon that included some 0603 bits that were difficult for him. looked at the BOM and saw nothing smaller than 0805 bits. He is concerned if HP-1 has any smaller than 0805.

 

The HP-1 has 10 components (out of 200+) that are smaller than 0805. Those are some 1N4148 diodes in SOD-323 packages. They're annoying. They're mostly annoying because the package has a slight slope to it, so you have to be careful that they don't go flying off the tweezers.

If your buddy just wants help soldering those diodes, he can slip me a $20 and I can solder the diodes for him before I ship the board. To make it cost-effective I'd have to combine the order for the diodes with another Mouser order so there may be a slight delay. I order from Mouser a couple of times a month, so the delay shouldn't be too long.

Just toss me an email (see Contact Us on my website) and we'll get it set up.

Tom