O2 amp CRC, diode, cap, and heatsink mods - Page 9 - diyAudio
Go Back   Home > Forums > Amplifiers > Headphone Systems

Headphone Systems Everything to do with Headphones

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 6th June 2012, 12:28 AM   #81
JZatopa is offline JZatopa  United States
diyAudio Member
 
Join Date: May 2003
Location: Chicago
Are there any problems with using the higher V batteries without making any other changes to the O2? I am guessing you would take longer to charge the batteries but I am not sure.
  Reply With Quote
Old 6th June 2012, 12:52 AM   #82
diyAudio Member
 
sofaspud's Avatar
 
Join Date: Nov 2005
Location: San Antonio
Any 9V-type battery will work without mods.
Battery charge time depends on state of discharge and capacity, so voltage isn't a concern.
__________________
It is error only, and not truth, that shrinks from enquiry. - Thomas Paine
  Reply With Quote
Old 6th June 2012, 01:35 AM   #83
JZatopa is offline JZatopa  United States
diyAudio Member
 
Join Date: May 2003
Location: Chicago
Quote:
Originally Posted by sofaspud View Post
Any 9V-type battery will work without mods.
Battery charge time depends on state of discharge and capacity, so voltage isn't a concern.
Even with the 9.6v not the 8.4v?
  Reply With Quote
Old 6th June 2012, 03:02 AM   #84
diyAudio Member
 
sofaspud's Avatar
 
Join Date: Nov 2005
Location: San Antonio
Yes.
Note that that voltage is still less than the ~11.5VDC that powers the O2 from the AC supply.
__________________
It is error only, and not truth, that shrinks from enquiry. - Thomas Paine
  Reply With Quote
Old 6th June 2012, 03:36 AM   #85
JZatopa is offline JZatopa  United States
diyAudio Member
 
Join Date: May 2003
Location: Chicago
Thanks sofaspud. Thats what I thought but I wanted to double check. I have little experience with NI-MH charging circuits and the last thing I want is to mess up the amp I just built.

I have enough parts to make a second amp and I think I am going to try the upgrade on the coupling caps that AGDR talks about in post 7. The sims look like it improves low end response and phase (if I am reading the graph right). This may not make an audible difference on all headphones but if I try using the amp for a few other purposes it might make a difference.
  Reply With Quote
Old 13th June 2012, 10:52 PM   #86
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
Default O2 current buffer mod for AKG K550s

This is a reply to a post by ethanolson in the main O2 thread here: The Objective2 (O2) Headphone Amp DIY Project. Since it involves modifications I figured best to post it here.

I ordered a pair of K550s a few days ago. Although I haven't received and tried them yet, just from the specs (SPL = 114dB/V) it looks like no voltage gain will be needed like you are showing - attenuation needed in fact with my sources - just potentially need some current buffering. So I'm in the process of stripping down an O2 for fun to be just a current buffer, specifically to work with the AKG K550s.

The first gain stage (U1, the NJM2068) is removed along with the 10K input resistors R14 and R20. The input signals on each channel are then run directly to the pot sections (which are also 10K, so the input RF filter still gets formed with C11 & C12 with the same corner frequency), which then allows signal attenuation. All of the gain and compensation parts are removed or disconnected as not needed here: R16,17,19,20,21,23; C19,20; S2. Then for the output stage current buffers I'm removing one of the 1R output resistors, R11 & R18, on each channel (so the two op amp sections are not in parallel anymore - only need 1/2 of each NJM4556 for the K550's low current requirements). Then jumpering the remaining 1R on each channel, R10 & R15, so there is O ohm output resistance (the 1Rs are not needed to balance anymore if the op amp sections are not in parallel). Then I'm going to drop the voltage rails to +/- 6Vdc (or maybe +/-5Vdc) with LM7806/LM7906 regulators and a 9VAC power transformer (or maybe can just use LM78L06/LM79L06, the 100mA TO92 versions) since the K550 should only need around 1.1Vpeak, absolute max.

For batteries I'm going to try using a couple of 3.6Vdc 600mAh NiMH wireless phone batteries. At least that is the plan right now, along with adjusting the power management circuit to work with the new lower rail voltages.

The net result is sort of a mini version of OPC's "Wire" headamp that is just a current buffer. Once I get it built and tested with the phones I'll post details in the mod thread. If someone gets something similar built first and tested, please post the results!

Congratulations again to RocketScientist for coming up with such a versatile design.

Last edited by agdr; 13th June 2012 at 11:19 PM.
  Reply With Quote
Old 14th June 2012, 04:47 AM   #87
diyAudio Member
 
Join Date: Nov 2011
Location: http://binged.it/ukmr5Y
This sounds really good to me for my IEMs and my hope of eventually getting the K550.

I thought about doing the same kind of thing, but your approach is more involved and more logical than my thoughts. For instance, I didn't consider removing the output resistors, but I did consider reducing their rating.
  Reply With Quote
Old 15th June 2012, 10:57 PM   #88
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
Quote:
Originally Posted by ethanolson View Post
I thought about doing the same kind of thing, but your approach is more involved and more logical than my thoughts. For instance, I didn't consider removing the output resistors, but I did consider reducing their rating.
I was glad to see your post in the O2 thread! I had worked out some numbers in my head that seemed to show that just various levels of attentuation would be needed with the AKG K550s, no voltage gain, and that is what you came up with too.

I ran some LT Spice simulations, below. The first concern with the modification is whether the single section of the NJM4556 can handle the current load of the 32R K550s at normal listening levels, or if the parallel buffers would be needed. A second concern is stability against oscillation in removing the 1R output resistors entirely with a typical headphone cable capacitive load. A third concern is whether the new +/-6Vdc power supply rails provide enough voltage headroom across the output devices in the NJM4556 when pushing 36mA(peak) or so.

My back of the envelope math seemed to indicate that a single section would do, since 1.1Vpeak going to the phones (which should be super loud as ethanolson mentioned given the K550s sensitivity) should be around 36mA(peak) [= 25mA(rms)]. Sure enough, that is what the first plot below shows. Green is the 1.1Vpeak input signal to the amp (and on to the phones since the "pot" is all the way up), blue the output from the 1:1 current buffer, and red the load current on the right axis. I still have 390pf across the 32R load to simulate some headphone cable capacitance.

The second plot is the frequency response. Still the same as the unmodified O2, as it should be, with the exception of a slightly higher RF 3dB point since RocketScientist had some compensation in the gain stage.

The third plot is an intentionally huge input signal to force clipping, to roughly test the accuracy of the device math model being used for the NJM4556. The output current for the single section limits at 70mA, exactly like it is supposed to, and the voltage overhead (output device minimum resistance in the chip) of the NJM4556 is about 4Vdc at 70mA (2Vdc across the phones = 6Vdc rails total). That is better than I was expecting. My back of the envelope came up with about 4V at the 36mA, which is why I picked +/-6Vdc rails. This would indicate +/-5Vdc rails may work for normal listening levels, but that is starting to split hairs and would have to be confirmed with some scope measurements. I'm leaving the rails at +/-6Vdc to allow some extra voltage headroom.

With the gain stage gone one slightly negative effect will the input impedance and upper (RF) corner frequency varying slightly with the pot position, since the 40.2k bias resistors on the output stage are in parallel (ac circuit model) with the pot wiper. Shouldn't make a bit a difference with any sort of normal source device, from what I can tell. On the plus side no forced clipping with large input signals, like my two sources that output 2Vdc, since the pot attentuates the input signal from 0.0->1.0 the standard way.

Input power to the NJM4556 chip would be around 25mA(rms) * 6V = 150mW. Subtract what the phones use per channel, (25mA)^2 * 32 = 20mW leaves about 130mW dissipated in the chip. Well within specs of 350mW per half (700mW total for the DIP8 chip). I've shown in previous posts you really have to plot a few points vs. output voltage level to find max chip dissipation, but this is far enough within specs it should be just fine.

Last edited by agdr; 15th June 2012 at 11:19 PM.
  Reply With Quote
Old 16th June 2012, 03:45 AM   #89
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
Talking O2 K550 current buffer mod is close to the Grado RA1

Well it just occurred to me that the net result of the O2 current buffer modification for the AKG K550s, above, is rather close to what is inside the Grado RA1 headphone amplifier that goes for $350.

Here: http://web.archive.org/web/200910271...RADO_schem.gif (wayback machine, give it 15 seconds)

The RA1 uses a single section of the NJM4556 with no output resistor, just as in this O2 mod. The RA1 has the NJM4556 wired for voltage gain though, where in the O2 it is just a unity gain current buffer. The RA1 runs on +/-9V while this mod runs on +/-6V. The RA1 uses a 100k pot and 5uF coupling cap, whereas RocketScientist's circuit uses a 10K pot and 2.2uF cap, which are just fine.

Last edited by agdr; 16th June 2012 at 03:48 AM.
  Reply With Quote
Old 16th June 2012, 09:22 PM   #90
diyAudio Member
 
Join Date: Nov 2004
Location: Germany
I once saw a RA-1 schematic using an inverting configuration... then the high-value feedback resistors make sense at least.

Essentially you're turning the O2 into the "cMoy with no gain" that RocketScientist reviewed at one point (well, a slightly better designed version anyway). If minimal current consumption is paramount and the source can easily drive a 10k pot, the mods as stated make sense - but do note that with only one set of buffers, distortion when driving low-impedance loads (K550s included) will be a factor of 2 worse.

It would be interesting to know how a single 4556 fares against the proposed low-power solution, 2x OPA2277.

How do you intend to "unparallel" the 4556 halves without PCB surgery? The two noninverting inputs are tied together after all...

Last edited by sgrossklass; 16th June 2012 at 09:28 PM.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Reduced diode switching by placing a cap across Goetz Power Supplies 10 19th December 2011 12:38 AM
DC blocker diode/cap orientation Glen B Solid State 4 22nd August 2011 05:46 PM
Mounting a small signal diode on heatsink - need suggestion IG81 Solid State 14 3rd November 2009 04:12 PM
diode bridge - 10000uF cap - diode tube - small cap - HV+ engels Tubes / Valves 5 29th January 2008 10:16 PM
Diode Heatsink dmk Tubes / Valves 8 9th September 2007 04:03 PM


New To Site? Need Help?

All times are GMT. The time now is 06:43 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2