Beautiful build vvs! I like the big filtered Schurter.
I received my Burson op-amps in the mail so dropped 'em into the unit. I had a single version before hence it wasn't working. The dual works perfectly well and sounds a little less constricted than the Texas Instruments op-amp that's in the BOM. The difference was much less pronounced than if you use a Burson in a DAC, but still an improvement and worth the price of admission (you only need one). I found the V5i picked up a little more noise in this amp, so stick with the V5 if you get one. I also swapped in and out the optional capacitors just in case they improved the sound with the Bursons, but wasn't able to hear any difference in any of the configurations.
Also... if you use metal-cased caps for the input, I found I needed to ground them, the signal is very delicate right there and can pick up noise.
I received my Burson op-amps in the mail so dropped 'em into the unit. I had a single version before hence it wasn't working. The dual works perfectly well and sounds a little less constricted than the Texas Instruments op-amp that's in the BOM. The difference was much less pronounced than if you use a Burson in a DAC, but still an improvement and worth the price of admission (you only need one). I found the V5i picked up a little more noise in this amp, so stick with the V5 if you get one. I also swapped in and out the optional capacitors just in case they improved the sound with the Bursons, but wasn't able to hear any difference in any of the configurations.
Also... if you use metal-cased caps for the input, I found I needed to ground them, the signal is very delicate right there and can pick up noise.
Last edited:
Hi Wayne, thanks so much for this awesome contribution to headphone DIY. The build is great as it's easy enough for a beginner but also a great sounding piece of kit for those of us with a few under our belt. The resistors are heat resistant Koa Speers 2W carbon film - also branded as the boutique Kiwame brand. They're pretty close to carbon comp in terms of warmth without the noise to worry about. I have a stash of them from when they were 10-16 cents a pop at Mouser, now they're more like 50 cents, or over a buck from boutique sellers.
...wouldn´t it be also possible to lower the supply voltage of the OP
by increasing R35/36 ?
the datasheets of all 3 Muses say iddle curent is 8-12mA. i don´t know
how much current the OS pulls from the OP, but it can´t be more
than a few mA´s, so even wit the original 47Ω of R35/36 voltage
would drop about 0,5V and with 150 - 200Ω you would be in the
safe Voltage range...
An Op amp is the last thing I need to add to my Mouser order.
Is this the correct LM833?
TIA
LM833N/NOPB Texas Instruments | Mouser United Kingdom
Is this the correct LM833?
TIA
LM833N/NOPB Texas Instruments | Mouser United Kingdom
Runs all my headphones effortlessly!
View attachment Whammy 1k.bmp
View attachment whammy 20k.bmp
I built this using the opa2604 and Fairchild FETS. I already had the case, transformer and many of the parts for the build. My 1uf Wima MKP10's fit like a charm. I have been listening to it for several days and I can't find any faults with this amp, extended nicely on both ends.
I added the waveforms of 1k and 20k and they look good to me. The DC offset is 1.5mv and .5mv again pretty darn good🙂
I was wondering why R39 and R40 are used? Are they used to help with the tracking of the ALPS pot? Also since I am using a stepped attenuator can I omit them? I only had RN65's at 2.2k and they are really big.
Anyway, thanks Wayne it is a nice build and I find myself being drawn to my computer to listen to music. "that's a good thing"
View attachment Whammy 1k.bmp
View attachment whammy 20k.bmp
I built this using the opa2604 and Fairchild FETS. I already had the case, transformer and many of the parts for the build. My 1uf Wima MKP10's fit like a charm. I have been listening to it for several days and I can't find any faults with this amp, extended nicely on both ends.
I added the waveforms of 1k and 20k and they look good to me. The DC offset is 1.5mv and .5mv again pretty darn good🙂
I was wondering why R39 and R40 are used? Are they used to help with the tracking of the ALPS pot? Also since I am using a stepped attenuator can I omit them? I only had RN65's at 2.2k and they are really big.
Anyway, thanks Wayne it is a nice build and I find myself being drawn to my computer to listen to music. "that's a good thing"
R39 and 40 insure you always have some real resistance in the circuit. In some situations it could oscillate and extremes of the pot. You may not need them.
First build here. 6L6 mentioned using other wire than the mic cable but does it still need to be shielded? Goal is to make this into a preamp as well with 3 in and 3 out so i will be using a lot of wire for all the switching.
Thanks!
Thanks!
I did not use shielded wire in my build. I twisted input and output cables as shown a few posts above.
I just finished building the headphone amp.
I am having fun testing different op amps.
One concern:
I found AD826 is extremely hot! The op amp top is over 45º and can't even touch it with my fingers.
Around the op amp base my IR thermometer indicates a scary 50º?!
The 3 neighbouring caps are about 44º.
Nonetheless, the AD826 sounds GREAT! It still works fine after 2 hours.
Is it normal or is overheating of the op amp dangerous?
I also tried LM4562 - a bit warm but below 33º;
RC4580 runs cool.
Thanks.
I am having fun testing different op amps.
One concern:
I found AD826 is extremely hot! The op amp top is over 45º and can't even touch it with my fingers.
Around the op amp base my IR thermometer indicates a scary 50º?!
The 3 neighbouring caps are about 44º.
Nonetheless, the AD826 sounds GREAT! It still works fine after 2 hours.
Is it normal or is overheating of the op amp dangerous?
I also tried LM4562 - a bit warm but below 33º;
RC4580 runs cool.
Thanks.
What voltage did you set your regs to? AD826 is specified to 5-15V (absolute maximally 18V according to spec sheet), so if you're running 17 or 20V that might be the explanation.
LM4562/RC4580 are specified to 17V/18V respectively, which could explain the difference in temperature...
LM4562/RC4580 are specified to 17V/18V respectively, which could explain the difference in temperature...
AD826 consumes up to 15mA for 2 channels without load.
Even with +/-15V, that equals 450mW.
Typical Rthja for DIP8 is 110K/W.
So it will mean 50°C above ambient.
https://www.maximintegrated.com/en/app-notes/index.mvp/id/3930
One solution is to stick a heatsink on top, e.g.
http://uk.farnell.com/fischer-elekt...x-10/led-heatsink-with-pins-square/dp/1850041
That has a Rth of about 30K/W.
Combined with the Rthjc of the DIP8 at 40K/W, you can reduce the temperature rise to 35°C.
At an ambient temperature inside your amp of say 30°C, that would be 65°C at the junction.
(I have not counted the thermal resistance of the thermal goue, which may add another 5°C or so.)
It will be about 45~50°C on the heatsink alone.
Patrick
Even with +/-15V, that equals 450mW.
Typical Rthja for DIP8 is 110K/W.
So it will mean 50°C above ambient.
https://www.maximintegrated.com/en/app-notes/index.mvp/id/3930
One solution is to stick a heatsink on top, e.g.
http://uk.farnell.com/fischer-elekt...x-10/led-heatsink-with-pins-square/dp/1850041
That has a Rth of about 30K/W.
Combined with the Rthjc of the DIP8 at 40K/W, you can reduce the temperature rise to 35°C.
At an ambient temperature inside your amp of say 30°C, that would be 65°C at the junction.
(I have not counted the thermal resistance of the thermal goue, which may add another 5°C or so.)
It will be about 45~50°C on the heatsink alone.
Patrick
Last edited:
Thank you Wayne. I have another amp on my workstation that has almost the same opamp input stage and it doesn't use the input resistors and I never noticed any problems.
I removed R39 and R40 and tested it with my signal generator and oscilloscope and experimented with it across a range of frequencies and adjusted the volume up and down and it didn't misbehave. I really don't think it will give me any problems as I usually set the volume to somewhere above mid-level and use foobar to adjust volume as I am listening.
Here is the link to the heatsink mentioned above at the manufacturer (Fischer ICK BGA 10 x 10 x 10) :
ICK BGA 10 x 10 x 10, Heatsinks for BGAs, Heatsinks f.cool, Fischer Elektronik
Patrick
ICK BGA 10 x 10 x 10, Heatsinks for BGAs, Heatsinks f.cool, Fischer Elektronik
Patrick
I did the configuration of LED+CAPs so it is about +/-16.8v.
I will turn it back to 15V and check again. I just like some indication that it is "on".
I already stick a tiny heatsink on top. Maybe too tiny. I will get the big one!
Thanks for the help!
I will turn it back to 15V and check again. I just like some indication that it is "on".
I already stick a tiny heatsink on top. Maybe too tiny. I will get the big one!
Thanks for the help!
The 1.8V difference only amounts to ~12% extra dissipation.
If you indeed want to use the AD826, then you need the heatsink for long term reliability.
Patrick
If you indeed want to use the AD826, then you need the heatsink for long term reliability.
Patrick
I went to the diyAudio store a few days ago and bought the (prepurchase) Amp Camp Amp and the whammy pcb.
The lm4562 will be first in the socket.
DT
The lm4562 will be first in the socket.
DT