[alsq]

I also posted this in amplification, but I got no answers. S'pose
this need be addressed to the DIY croud then.

I listen to two META42, one with the AD8610, the other with
OPA2132. Would there be any improvement in a PPA or a
Gilmore? I am quite happy with my current amps; I use the
AD8610 at 12-14V with a linear power supply on Sennheiser
HD580, and the OPA2132 with one 9V battery as a *portable*
with Sony EX71 earbuds. I just ordered a set of AKG
K240S, too, so I guess I may need another amp and may
then change the can/amp matchup. So: it's not critical that the
(possible) next amp be portable; but would I be able
to tell the difference? The PPA uses AD8610, too; the
circuit topology is different, but it is also broken up into distinct
voltage/current amplification section, so why should it sound
better? The Gilmore is discrete and class A, which makes
it a very different animal, but is different also better?
I don't feel like blowing 2-300$ and an inordinate amount
of DIY time (well, that's fun, though, except I'm really busy)
into another amp that also sounds well, but then I can't tell
the difference from my current setup.

I heard the Senns do change when hooked to improved
quality amps, though. What is one to do?

 

[Karlosak]

I'm in the same situation, torn between PPA and Gilmore as the next DIY project. So I would really like to see an answer.

Come on guys! Don't tell me there is nobody who has at least heard or made a comparison of PPA and Gilmore!?!

 

[AIM9x]

I kind of skipped over the whole cmoy/meta/a47 thing and went straight for gilmore.


top
bottom


still have a bit of testing to do... the PS is acting up.

 

[alsq]

aim9x

Nice pics. The jumpers on the component side,
those btw. the LEDs, that cross each other--are
they supposed to be insulated from each other?

What PS are you using? Some PSUs built for
Gilmore are so outrageous

I wonder if they
deserve the effort. Maybe an adjustable regulator
like a LM3X7 pair in a sensible (i.e. non trivial)
configuration would just do a job good
enough, and the difference with a tracking
PS would be mostly hypothetical.

 

[AIM9x]

that picture is missing two 3.3k resistors, but other than that, it's complete. I need to switch out the LED on the left channel due to too heavy of a DC offset.

All the wires that overlap in the pictures have a good 2.5-3mm of clearance.


as for the PS... I am using a gilmore spec PS at the moment, but the OPA548 on the positive rail is giving me some problems (negative rail's opa548 is working perfectly). I may simply remove the opa548 from the positive rail and reset the gain on the PS later and see what results that will give me.

 

[JohnFerrier]

From "A Pure Class A Dynamic Headphone Amplifier" by Kevin Gilmore

"The amplifier's output is voltage limited (not current limited) because the second stage runs out of voltage swing - typically about 6V rms. For a 32 ohm load, the maximum output power is 1.125W (0.1875A). For a 300 ohm load such as the Sennheiser HD600, the maximum output power is 0.125W (0.02A). To increase the maximum output voltage to, say, 10V rms, try increasing the power supply to ±20VDC and change the 500 ohm resistors in the current sources/sinks to 600 ohms (or ±24VDC and 700 ohms - but careful not to fry the output transistors)."


Something to keep in mind when building Gilmore's amplifier. I don't know what people's experience is using Gilmore's amplifier to drive Sennheisers loud, but some might think that this is a bit light. Of course, building the bridged version provides more voltage and current. I recommend the bridged version.


JF

 

[AIM9x]

Quote:

Originally posted by JohnFerrier I don't know what people's experience is using Gilmore's amplifier to drive Sennheisers loud, but some might think that this is a bit light. Of course, building the bridged version provides more voltage and current. I recommend the bridged version.

I don't know where you people come up with this junk. The Gilmore Dynamic does not have a problem "driving sennheisers loud." Unless you've had both your ears ground to a pulp and fed to Albanian snapping turtles while you were a prison of war in Greenland, you're not going to have a problem driving Senn hd580/hd600 with the gilmore.

Unless you want to replace your headphone's cables as well, don't bother with the balanced bridge version.

 

[JohnFerrier]

Quote:

Originally posted by AIM9x I don't know where you people come up with this junk.

I'm not claiming that it is not loud enough (personally, I enjoy classical music at moderate levels). However, having carefully read Kevin's description (a recommended reading with a lot of useful information), I noted that Kevin concedes that the unbridged version drives Sennheisers to 125mW (see previous quotation). Now I am unable to find the maximum rating of the Sennheiser HD-580/600s, but likely the amplifier will not drive them to maximum sound levels. By my calculation the unbridged version will drive the HD-600s to about 118db.

Someone correct me, if I'm wrong:
HD-600 sensitivity: 97db @ 1mW (~547mV)
Gilmore output: 125mW (~6.123Vrms)

6.2Vrms / 0.55mV = 11.3 (~21db)

Max. HD-600 SPL = 97db + 21db = 118db

I read somewhere that the HD-600 are capable of 122db.

Obviously, 118db is loud. The only point is that the amplifier will clip, before the headphones reach maximum capacity. (The Sennheisers are the wrong headphones to use to listen to extremely loud music anyways--like a 130db Who concert.) Some "audiophiles" prefer amplifiers that are capable of overdriving the transducers (not that they actually listen to it that loud).

This uses the 1khz specifications. It may be worse, or better, at other frequencies.

Quote:

Originally posted by AIM9x Unless you want to replace your headphone's cables as well, don't bother with the balanced bridge version.

Even if the unbridged version drives the HD-600s to full capacity, I still recommend the bridged version. Kevin makes it easy, in that the output transistors are driven by the reverse sides of the input. Nice. (And, if you know what you are doing, you don't have to buy replacement cable to drive the Sennheisers differentially. However, you do need to cut the cable and add a jack. Did I mention that you need to know what you are doing?)

I have one question regarding the Gilmore amplifier. I see in the picture he has a knob to control the volume. What value is recommended 10k, 50k???

And, if the likes of Kevin Gilmore are responding, don't discrete FET input amplifiers have the same problem with distortion caused by the varying input capacitance versus voltage that integrated op amps do? Noticing that the 2SK389s have 25pF input capacitance, I wonder if the discrete topology avoids this or not. (Discussion of "Source impedence and distortion" found in OPA134 datasheet and other datasheets.) This is the reason people use the Jung multi-loop feedback on op-amps. Maybe a low impedence attenuator is recommended. (I'm thinking of a low impedence attentuator rather than the additional noise that the multi-loop components add.) The reason I ask is I'm (as others) trying to find the very best topology to build. Currently, the leading candidate is a cross-coupled differential amplifier using op-amps all biased into class A (capable of overdriving the HD-600s to 127db).

Regarding The Who at 130db, additional distortion doesn’t necessarily ruin the sound experience (though this is only my best guess--someone correct me, if I'm wrong).

JF

 

[alsq]

What do the likes of Kevin Gilmore think of the YoungKyoo Kim thermally compensated mod?
I don't have a Gilmore, but I may be building one,
and I wonder, I wonder

 

[JohnFerrier]

Quote:

Originally posted by alsq What do the likes of Kevin Gilmore think of the YoungKyoo Kim thermally compensated mod? I don't have a Gilmore, but I may be building one, and I wonder, I wonder

Could someone post a link to find information regarding?


JF

 

[Glassman]

http://www4.head-fi.org/forums/showt...s&pagenumber=1

scroll down..


as for Kevin's opinion about it:
Quote:

I have not tried the modified variant. I'm not absolutely sure it stays 100% class A throught the entire voltage range. The effects are minor to be sure.

 

[JohnFerrier]

Quote:

Originally posted by alsq What do the likes of Kevin Gilmore think of the YoungKyoo Kim thermally compensated mod? I don't have a Gilmore, but I may be building one, and I wonder, I wonder

I'm not the likes of Kevin Gilmore (to be sure, I'm not sure that anyone is the likes of anyone else). I'm not sure that Kevin is actively following this thread.

Without studing what YoungKyoo did, it seems like he did make small improvements. For example, I think that the thermally bonded diodes help with temperature management. Using values that match 1.85V LED aids DYI'ers, too. Note in Kevin's reply that he didn't say that it ruined his design and it won't work. With any design, there are parameters that can be changed without causing problems (and possibly improving things). I don't think that YoungKyoo would go to this trouble to make changes that make it a worse circuit. Another set of eyes adds to the collective "wisdom". I recommend his changes.

As far as falling out of class A, I don't know. However, Kevin describes how even with the original design, it can fall out of class A.


JF

 

[tangent]

Quote:

Obviously, 118db is loud.

Yes. It's basically at the threshold of pain. If you insist that even approaching this SPL is necessary, I assure you that in a few years the distinction between a Boostaroo and a Gilmore will be moot to you.

 

[JohnFerrier]

Quote:

Originally posted by tangent Yes. It's basically at the threshold of pain. If you insist that even approaching this SPL is necessary, I assure you that in a few years the distinction between a Boostaroo and a Gilmore will be moot to you.

Approaching it from which direction? Either way, thanks for your concern.

I'm interested in a design that can readily drive high current requiring phones like the Sony CD3000 or high voltage requiring phones like the AKG K1000. (If it also happens to be able to drive my 4 ohm Yamahas, that is okay too.) However, it rules out the Gilmore, the Boostaroo, and the META42. (The Gilmore ought to be okay with the Sonys. And a bridged Gilmore might be okay with the AKGs--the YoungKyoo Kim thermal mods will help.)



JF

 

[AIM9x]

118dB... the senns would practically be loudspeakers. I don't see why anybody would even bother going over 100dB... let alone nearly 4x that loud.