JLH Headphone Amp - Page 6 - 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 2nd March 2010, 03:22 AM   #51
diyAudio Member
 
ashok's Avatar
 
Join Date: Jun 2002
Location: 3RS
I think you might have to look elsewhere. Maybe even the power supply ?
In my sim of the circuit 33pF does not show up any problems . All transistors are BC5XX and BD136 output . I didn't really think the 33pF caused the problem as it would just oscillate and not whistle with a variable frequency like it is doing now. It's like a signal generator , sweeping over the full audio band !

It's taking me a long time to do the board layout and its only 90% done. Hope I get it done today. It has an on board psu.
__________________
AM
  Reply With Quote
Old 4th March 2010, 02:52 AM   #52
diyAudio Member
 
ashok's Avatar
 
Join Date: Jun 2002
Location: 3RS
Hi Melorin,
Did you solve the problem ? I find in the sim that the square wave response is unstable with the component values you have. So maybe it's oscillating. The feedback cap and the zobel keep the amp stable with a capacitive load. I used 1000pF (parallel with 32 ohms as load ) for a long ( shielded ) cable. Many cables here are about 120pF per meter. So component values need to be picked for a stable operating point.
My headphone amp uses slightly different component values and the ones I have may not be as suitable for you. However my feedback from output to the input transistor has a 3.3K ohm resistor parallel with a series 10 ohm and 47 pF. The Zobel is 22 ohms with 0.1uF . Some component values I selected because that's what I have right now !

My board is finally ready , BUT I discovered that I etched it with a reversed pattern .....not again !
Luckily the transistors can be reversed. However the heatsink will now have to be within the board unless I flatten the power devices. So until I have some heatsinking I cannot power it up. Today is a busy day. Hope I get some time later this evening.
__________________
AM
  Reply With Quote
Old 4th March 2010, 01:17 PM   #53
melorin is offline melorin  Hungary
diyAudio Member
 
Join Date: Feb 2010
Location: Lőrinci
Quote:
Originally Posted by ashok View Post
Hi Melorin,
Did you solve the problem ? I find in the sim that the square wave response is unstable with the component values you have. So maybe it's oscillating. The feedback cap and the zobel keep the amp stable with a capacitive load. I used 1000pF (parallel with 32 ohms as load ) for a long ( shielded ) cable. Many cables here are about 120pF per meter. So component values need to be picked for a stable operating point.
My headphone amp uses slightly different component values and the ones I have may not be as suitable for you. However my feedback from output to the input transistor has a 3.3K ohm resistor parallel with a series 10 ohm and 47 pF. The Zobel is 22 ohms with 0.1uF . Some component values I selected because that's what I have right now !

My board is finally ready , BUT I discovered that I etched it with a reversed pattern .....not again !
Luckily the transistors can be reversed. However the heatsink will now have to be within the board unless I flatten the power devices. So until I have some heatsinking I cannot power it up. Today is a busy day. Hope I get some time later this evening.
I've given up. I have no more idea how to shut down the oscillation.
Furthermore, I have a new guest in my amp beside oscillation, low level hum, like the transformator would be heard, but the power supply is very good and is seperated from the amplifier.
  Reply With Quote
Old 4th March 2010, 05:15 PM   #54
diyAudio Member
 
ashok's Avatar
 
Join Date: Jun 2002
Location: 3RS
OK I finally finished the board and assembly. One channel seems to have a problem. I'm sure I made a mistake somewhere becaused I reversed the board layout ! I'll fix it tomorrow.
However the other channel works. I'm able to set the dc output to around +/- 1mV . The meter reads 0.000 V . It didn't change for about 1 hour that the amp was on. Will have to look at this closely tomorrow. I did set the dc offset after a 30 mins warm up period.

Frequency response was flat ( 0dB ) at 20 Khz and -0.2db at 20 Hz with a 33 ohm load. Max output ( 1% dist ) was about 3 Volts rms into 33 ohms.
Max output into 10 K ohms load was about 7 volts rms( less than 0.1% dist).
Second harmonic was at -74.2 dB and 3rd harmonic at -75.8 dB. All other harmonics were at -95dB or below.

Pity the second channel had a problem. I couldn't listen to it. Note that I'm getting a higher level of 3rd harmonic distortion now. The single supply unit had practically only second harmonic distortion. Must try to see why that is so. The noise floor is also very low and you can hardly see any power supply harmonics ! So there should be no hum on the can's.
The heat sinks do heat up but I didn't measure the temperature. Quiescent current appears to be at about 0.1A in the output stage.
Looks good so far. Will have to do a listening comparison with the single supply unit also.
Attached Images
File Type: jpg JLH split supply.jpg (63.0 KB, 1163 views)
__________________
AM

Last edited by ashok; 4th March 2010 at 05:32 PM.
  Reply With Quote
Old 4th March 2010, 05:24 PM   #55
diyAudio Member
 
Join Date: Nov 2007
Location: Dallas
Class A from the perspectve of a single supply, current draw appears uniform.
Class A with two supplies, each supply will see equal and opposite copies of
the output current swing.
  Reply With Quote
Old 5th March 2010, 03:07 AM   #56
diyAudio Member
 
ashok's Avatar
 
Join Date: Jun 2002
Location: 3RS
Thanks Kenpeter.
I will be doing the comparison test ( between single supply and split supply )sometime this weekend I think.
I didn't mention that the output impedance was less than 0.25 ohms and since there was a series resistor of 2.2 ohms the Zout is about 2.45 ohms. No problem for most headphones. Response errors due to varying headphone impedance should be under about 0.25dB.The 2.2 ohms is for stability under capacitive loading. You could reduce it to 1 ohm if required and check for stability.

If anyone wants to see the actual performance graphs, please tell me now. I could put them up. If no one is interested in them I'll just mention my findings regarding the sound and close this thread. Need to move on to other things !
Cheers.
__________________
AM
  Reply With Quote
Old 5th March 2010, 04:21 AM   #57
melorin is offline melorin  Hungary
diyAudio Member
 
Join Date: Feb 2010
Location: Lőrinci
Quote:
Originally Posted by ashok View Post
I'm able to set the dc output to around +/- 1mV . The meter reads 0.000 V . It didn't change for about 1 hour that the amp was on. Will have to look at this closely tomorrow. I did set the dc offset after a 30 mins warm up period.
Which version did you build? Is the potmeter for the 0 DC instead of R4? Could you show us the schematic of power supply? I try to find out why I have hum.
Could it be the problem that I have +11,7V and -12,1 voltages at the power output?
  Reply With Quote
Old 5th March 2010, 12:46 PM   #58
diyAudio Member
 
ashok's Avatar
 
Join Date: Jun 2002
Location: 3RS
Quote:
Originally Posted by melorin View Post
Which version did you build?

The one with the split supplies. The red LED is used in place of D3 / D4.

...Is the potmeter for the 0 DC instead of R4?

Yes the potmeter is used in place of R4. I used a 10 turn preset as I had that.
10 K ohms.


Could you show us the schematic of power supply?

I'll describe the power supply as it's easier than drawing it out now.
The transformer with 12 - 0 - 12 V secondary is connected to a bridge made up of four FR107 diodes. The positive rail feeds a 2,200uF capacitor. This is connected to the next 2,200uF cap via a 2.2 ohm resistor ( 1 watt). That filters a bit of the psu noise. This is followed by a LM7812 which has a 47uF capacitor at its output. The negative rectified rail is connected to a similar circuit using a LM7912 to give -12 V.


Could it be the problem that I have +11,7V and -12,1 voltages at the power output?
The voltages I have are + 11.8 V and -12.05 V after the LM regulators.

Are you using the same values for all parts as shown in the circuit diagram you posted?
__________________
AM
  Reply With Quote
Old 5th March 2010, 03:27 PM   #59
diyAudio Member
 
ashok's Avatar
 
Join Date: Jun 2002
Location: 3RS
Default The listening test !

I finally got to hear the new phones amp with the split supply. It sounds great ! The two main headphones I used were the Sennheiser HD580 and the Philips SHP 8900.
The HD850 sounds less bright than the 8900 but transient drums sound much tighter and powerful on it. The 8900 does go a bit lower than the HD580 and it's just about audible on some tracks. The HD850 sounds dull if you go from the 8900 to the 580. But the HD580 probably sounds smoother in the mid range. The image however is a bit flat. The 8900 sounds spacious !

I compared the single supply amp with the split supply amp. They sound quite similar. Especially with the Sennheiser ( 600 ohms). With the 8900 ( 32 ohms) the low end is slightly lighter but not very easily noticeable on the single supply amp. Transients are just as tight on both amps and I couldn't hear much of a difference in the mid range or HF. No audible hum or hiss on either amp.

The single supply version is safer than the split supply version because you have a ( usually horrendous!) electrolytic capacitor in series with the signal path. However unlike many locations this one is biased at a high 6 Volts dc and so maybe performs better than no bias or low biased elcos.
Main thing is that a crashed amp will not burn up an "expensive" pair of headphones. No series relay in the low current signal path and no dc servo which is frowned upon by many.

The split supply comes out ( slightly) ahead just because there is no capacitor in the output and just about noticeable on the 8900's 32 ohms load. DC offset shift drifted by about 1mV either way over a 2 hour period. Switch on from cold produced about 25 mV offset which settled down to about 4mV in about 2 minutes and 1mV in 5 minutes.
I would still prefer to have a shorting relay ( to ground) in case of dc offset error with an opamp to detect a dc offset of more than 5mV or 10 mV(?). That would protect the "expensive" can from going up in smoke in case the power transistors fail ( very unlikely I'd say but then Murphy is always lurking around the corner !)

If anyone wants any clarifications etc I'd be happy to reply.

Apart from that I'm ending this thread. I completed what I was out to find out and hope it helped all those who have been following the progress.

Melorin please contact me via PM .
__________________
AM
  Reply With Quote
Old 5th March 2010, 06:18 PM   #60
diyAudio Member
 
ashok's Avatar
 
Join Date: Jun 2002
Location: 3RS
I sat down and listened to several albums after my last post. The amp sounds really very nice. So if given a choice build the split supply one.
Maybe you could add a parallel relay protection if you really plan to use it all the time and with expensive headphones.
On switch on you do get a 0.5 volt surge which lasts a couple of seconds. Switch off also produces a surge. Maybe the phones need protection at that time ? Even if you plug the phones in after a while and remove it before power down, you could forget sometimes !
I'll check the dc surge on my scope later on.
Certainly sounds better than a power boosted opamp circuit using OPA2134 and TO126 type power transistors. I'm really glad that I tried out this amp. It just happened because of the new 'Headphones' section !
__________________
AM
  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
JLH headphone amplifier steveh49 Headphone Systems 8 14th May 2014 04:31 PM
JLH Headphone amp aethelbert Headphone Systems 4 14th July 2009 12:14 AM
JLH headphone amp with FET's and servo nelsonvandal Headphone Systems 17 30th August 2008 10:26 PM
PCB for JLH amp FaTTy Solid State 0 27th March 2007 05:40 PM


New To Site? Need Help?

All times are GMT. The time now is 04:38 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