Simply because one makes a statement in error does not make it an article of faith.
Thanks for pointing that out, Avro Arrow.
While I have been shown to be wrong in detail, the original point remains true, neither feedback nor paralleling are necessary to achieve the required output impedance.
That is indeed correct. Article of faith <> faith-based claim.
Quite so. Feedback though allows the desired output impedance to be attained with lower bias currents.
I'm content to explore the epistemological basis of your erroneous claim. What evidence did you base your 'pretty sure' on?
From BUF634 Application notes:
"The output resistance of the BUF634 is about 10 ohms.
Therefore, series output resistors for decoupling the
individual buffers are no longer necessary".
No faith need. Just fact.
"The output resistance of the BUF634 is about 10 ohms.
Therefore, series output resistors for decoupling the
individual buffers are no longer necessary".
No faith need. Just fact.
Faith-based, that would be received belief that I would cling to dogmatically, so I would find it difficult to say, for example, "Less than 1R? No, I was wrong. It's more like 10R".
No, I guess it was based on a misplaced faith in my memory.
No, I guess it was based on a misplaced faith in my memory.
'Cling to dogmatically' ? Not essential in my meaning of faith-based, no though certainly a possibility. Perhaps you're reading too much into the remark?
Last edited:
I'm not a technical guy, but I assume your guys are debating the merits or myths that is 'descrete' devices......all I know is that it's not easy to find uncolored and accurate sounding hifi gear. but I do like the shureV15III. It's a good start..if I could ony find a 1 ohm headophone output to plug it into. The 02 is what i'll try I suppose...unless anyone feels the extra chips are necessarily nasty to put the signal throguh. I do like an unpprocessed sound. which is why the buffer appeals to my sensabilities. I've no proof that either or both aren't necessarily possibly good enough, as long as their well done?
Last edited:
Back on topic, how about this for low/no gain:
DIY IRF610 MOSFET Class-A Headphone Amplifier Project
Thorsten recommended something like this as simple and sweet and I've got the parts, gonna give it a try.
Maybe add a JFET to buffer the higher input capacitance of the MOSFET a la Pass ...
Cheers,
Jeff
DIY IRF610 MOSFET Class-A Headphone Amplifier Project
Thorsten recommended something like this as simple and sweet and I've got the parts, gonna give it a try.
Maybe add a JFET to buffer the higher input capacitance of the MOSFET a la Pass ...
Cheers,
Jeff
Last edited:
It doesn't have a low enough output impedance for the OP's requirements. Changing the MOSFET from the relatively stone-age IRF610 to something more recent will almost certainly lower the Zout (higher gm).
Get any buffer + input opamp combo, wrap the buffer with feedback.
You won't get a good sounding amp with the buffer alone, as they usually have awkward input impedance/distortion figures.
Or check out LH0063, it is beasty, has fet input and hefty of current, without any sort of global feedback. 1R of typical output impedance in the datasheet... It is high-speed device, so put bypasses and lay out the GND accordingly.
You won't get a good sounding amp with the buffer alone, as they usually have awkward input impedance/distortion figures.
Or check out LH0063, it is beasty, has fet input and hefty of current, without any sort of global feedback. 1R of typical output impedance in the datasheet... It is high-speed device, so put bypasses and lay out the GND accordingly.
What is the output Z? Any rec's for a replacement MOSFET or other for the 610/lm317 buff ?
Last edited:
I've not spent very long looking for higher gm FETs so far, but here's one which looks better on paper than the IRF610 in this application. Attached Vgs vs ID plot - its an IRLR024. Notice at 100mA ID it needs just 0.25V increase on the gate to get 500mA, so gm is close to 2. IRF610 is specified at 1.25A for gm, a drain current much higher than the application demands. All is not a bed of roses with this part though because the reverse transfer capacitance has doubled (to 53pF).
Output Z means output impedance.
Output Z means output impedance.
Attachments
You
No, it's not a problem.
Headphone Amplifiers - Benchmark Media Systems, Inc.
All you have to do is pay the price.
Or you can accept the testimonial of the designer of the O2 regarding it's performance relative to the Benchmark.
No, it's not a problem.
Headphone Amplifiers - Benchmark Media Systems, Inc.
All you have to do is pay the price.
Or you can accept the testimonial of the designer of the O2 regarding it's performance relative to the Benchmark.
Last edited:
I just bought an 02 from JDS. I'll report back to testify if the extra chips can be heard or if it sounds totally natural. I suspect it will sound great.
I really am at the end of the rope over here. I received the 02 amp and it did not do it for me. I listen to vinyl and even this scenario sounds a bit hard and cold to me. the bass is also very un-tunefull and innacurate imo.
I need a bit of distortion or something because this is not right to my ears. BUT please do not suggest mushy class A distortionboxes. I just need a bit of something nothing outrageous. a class AB?
So considering this new piece of info, could anyone suggest again a modest schematic for a headphone out that will accomodate 12 ohm IEM(1 output imp or less) that is full bodied but natural sounding? thanks
I need a bit of distortion or something because this is not right to my ears. BUT please do not suggest mushy class A distortionboxes. I just need a bit of something nothing outrageous. a class AB?
So considering this new piece of info, could anyone suggest again a modest schematic for a headphone out that will accomodate 12 ohm IEM(1 output imp or less) that is full bodied but natural sounding? thanks
Last edited:
I'm sure its problems lie in the power supply - it was designed by the numbers. So if you supply nice clean low impedance power to it (an array of caps across the rails) it'll deliver more of what you want.
Im just using a standard wall suppy so I'm sure you are right, but still, I don't want to mess with this. I'd rather build something different from scratch that doesn't need such a robust power supply.
You'll need a design like ALD posted (post 29) but with the load taken across the transistor, not across the CCS (LM317 in that schematic). Then it'll be largely PSU independent.
- Status
- Not open for further replies.