On HD650s and speaker amps...

[Dusty Chalk]

I want to say I heard Hirsch's rendition once -- stevieo owns it now, so if we can get him to actually post, he can talk about its sound quality.

I have the ASL unit, and was pretty much underwhelmed. Perhaps I should open it up and take pics? I suspect the sound quality is entirely dependent on the quality of the transformer.

[ting.mike]

I'm thinking of making this adapter, and was wondering about the resistor values. If my physics isn't too off (it usually is...), do I understand from the 20 and 2 ohm resistor arrangement that the amplifier will 'see' a 2 ohm purely resistive load, and that the vast majority of the current will take the 'easy' route and be sunk straight to ground?

Would one not then want to adjust the value and relationship between the resistors to reflect firstly the ideal load for the amplifier, and secondly the amount of attenuation required to match the phones' sensitivity?

If my physics isn't too off myself, I think you're looking at the current flow in reverse. Again from the Grado picture, the current flows from the bottom to the top, so the amp will see a 20 ohms first, then the parallel combination of 2 ohm and the headphone impedance. But I'm no expert in electronics either.

As nate said:

A more informed opinion is definitely needed.

Actually the amp sees the 20 ohm load in series with the paralleled load of the 2 ohm resistor and the headphone. In the case of Grados this would work out to a bit under 2 ohms in series with 20 ohms.

I've built one of these doohickeys with 5W resistors way back in the day, my sources & 'phones weren't that great back then so I didn't really notice much of a difference, if any, between the adaptor output and the headphone output on my receiver.

This sounds more correct. Looks like the resistors in the Grado diagram is not an impedance-matcing device rather a level attenuator device. Besides, doesn't impedance varies with frequency? so, you can't impedance match a headphone to an amp using a resistor.

Cool ting.mike, Smeggy, and I were having a conversation regarding this late lat night in chat.

Wanted to verify one thing. The consensus last night was re-terminating my hd800's to speaker/banana plugs and using the speaker amplifier section of my b22 wouldn't yield any extra power to the headphones vs. the headphone jack?

Yes. Smeggy said that your b22 speaker terminal and headphone jack is the same output.

I have built a simple version which uses an 8 ohm 5+W resistor to load the amp and a 100R in series with the headphones

Rod Elliot has a more thorough approach to balance power supply and impedance.

Thanks, I'll read the Rod Elliot article.

..dB

[ting.mike]

Just read this post by Zorander on HF:

Actually, if the phones are terminated with a plug with common ground (i.e. 1/4" TRS), you can't just put an adapter to it and not risk bridging the speaker taps. It's better to completely snip off the plug and re-terminate as appropriate.

Ack! Is this for real? I'm doing exactly that-- using a 1/4 TRS jack and bridging the speaker taps!

[TheSloth]

Just read this post by Zorander on HF:

Ack! Is this for real? I'm doing exactly that-- using a 1/4 TRS jack and bridging the speaker taps!

You are tying the grounds together, so with a balanced amp, yes you'd be causing big problems. With single ended I can't see how it's an issue - the grounds are usually tied together anyway.

[n_maher]

There are a couple of bottom lines here:

1. If you don't know and don't know how to check how your speaker amp's outputs are configured you shouldn't be messing with it. The Grado adapter picture clearly shows 4 conductors so that it doesn't have to worry whether or not the output of the amp has a common ground connection.

2. It's pretty likely that the Grado adapter had its resistor values chosen for a Grado headphones (all of which are 32ohm). Assuming that it'd be either appropriate or useful with headphones of varying impedances is a "at your own risk/peril" type of deal. It may work great, it may load your amp in such a way as to cause it to go into thermonuclear meltdown.

Me, I'm content to have a separate speaker and headphone amplifier.

[ting.mike]

There are a couple of bottom lines here:

1. If you don't know and don't know how to check how your speaker amp's outputs are configured you shouldn't be messing with it. The Grado adapter picture clearly shows 4 conductors so that it doesn't have to worry whether or not the output of the amp has a common ground connection.

2. It's pretty likely that the Grado adapter had its resistor values chosen for a Grado headphones (all of which are 32ohm). Assuming that it'd be either appropriate or useful with headphones of varying impedances is a "at your own risk/peril" type of deal. It may work great, it may load your amp in such a way as to cause it to go into thermonuclear meltdown.

Me, I'm content to have a separate speaker and headphone amplifier.

Thanks, nate.

I just talked on the phone with a local based headphone amplifier designer.. and he shared some insights: (retyped & paraphrased based on my understanding)

1. No problem with sharing the ground on the 1/4 TRS, as on unbalanced config they are always shared. Then I notice that Grado's adaptor also use 1/4 TRS female jack at the end.

2. The impedance matching device is useful if the speaker amplifier has limited power/current, as matching the impedance will allow the amplifier to produce more current. As I'm using a 300W mosfet, I have no use for such devices.

3. The Grado Adaptor:

This is how it works (according to him)

- The 2 ohm resistor is in parallel with whatever headphone you plug in. And regardless if you use a 32 ohm Grado or 300 ohm Senn, the resulting impedance will be somewhere less than 2 ohm (1.8~1.9 ohm) So, I suppose this design is not limited to only Grado headphones.

- This is where the 20 ohm resistor comes -- to bring that ~2 ohm resistor back to safer levels (don't want the amp driving a 2 ohm load into headphones )

- On choosing the 20 ohm resistor, he said that Grado probably tested different values and find out what sounded best. Generally, he said, if you increase the impedance, the bass will be greater. (Hmm... add 200 ohm to the AKG 501 for bass boost?? )

Okay, hope that is all good now.

Edited June 28, 2009 by ting.mike
Added information

[TheSloth]

Me, I'm content to have a separate speaker and headphone amplifier.

I agree with you there, but I don't trust speakers as evaluative tools for audio gear. I'd like to be able to plug my headphones into my power amp to get an idea of how it performs.

[TheSloth]

- This is where the 20 ohm resistor comes -- to bring that ~2 ohm resistor back to safer levels (don't want the amp driving a 2 ohm load into headphones )

- On choosing the 20 ohm resistor, he said that Grado probably tested different values and find out what sounded best. Generally, he said, if you increase the impedance, the bass will be greater. (Hmm... add 200 ohm to the AKG 501 for bass boost?? )

What does all of this resistorage do to the output impedance of the amp?

[ting.mike]

What does all of this resistorage do to the output impedance of the amp?

It wouldn't do anything to the amp's output impedance. It just "changes" the headphone impedance -- so the amp sees ~22 ohms instead of 32 ohms. If my understanding is correct.

[ting.mike]

Had a mini meet today with a Grado guy who doesn't like the HD650, and we both listened to the HD650 through the Adcom Power Amplifier again, and we were really impressed with:

1. Speed

2. Staging

The problem was that my Adcom is quite noisy when plugged into the HD650, which is a big pity, cause otherwise the sound is just way too good.

Seriously more HD650 owners should give this a try. I don't know what headphone amp can add speed to the HD650.

[ting.mike]

Ok, so I got a plain wire (Kimber 4TC speaker cable), put Sennheiser Cardas connector on one end, and plug the HD650 straight to the speaker terminals. Compared to using 1/4TRS jack (shared ground), this eliminates a LOT of noise. Only very small noise is left. Turns out most of the noise is from the pre-amp (Adcom GFP 565), while the power amp has noise, but very little (GFA 5802).

I tried adding a resistor to kill the noise, and while it worked, the resistor I happen to have is too big (470 ohm), and it changed the HD650's frequency response a lot, making it bassy and veiled, so I'm getting a smaller value resistor only to kill the noise (perhaps 47 ohm). More on that later.

I'm also contemplating building a passive pre amp now, since the source is a balanced source, and the power amp is a balanced power amp. I figured it's basically a box with XLR jacks in/out, and a Potentiometer, shouldn't cost that much. And I'll be able to eliminate the weakest link in the chain (preamp GFP 565) which is unbalanced.

I hope someone else is also finding great results powering their HD650 from speaker amps..

[Smeggy]

Isn't this all similar to a zobel? 20-22 ohm 2W resistor and small cap.