Using the AD844 as an I/V

[georgehifi]

Pic of the temp setup for those who wish to see.

Cheers George

 

[Torchwood421]

BUF03

Got the BUF-03 yesterday put it in instead of the AD844's own buffer, dc offsets steady at .1 to .2mV after warm up, when cold I get around 5mV, also a mute circuit is needed as it in conjunction with the AD844 gives a hefty 5v switch on and off thump for a second. I have very nice TO-99 cast aluminium heatsinks for them they run HOT, and no they are not oscillating, 15v rails.
The 844's buffer will not drive my Sennheiser HD650 (300ohm) without gross distortion where the BUF-03 does it no trouble, the BUF-03 has far lower output impedance and higher current than the 844's buffer.
On my system proper both are driving a 50kohm total load of the input of my poweramps (bi-amped 100kohm each) I use the digital domain volume on the cdp, it is in the same position for both buffers almost full up.

Listened all night. I am perplexed, the sound is so different and opposite to what I thought would happen, the AD844 is super fast, very dynamic, very detailed, sweet highs, very extended both ends and kicks like a mule in the bass, and has subterranean bass. It can get in your face sometimes with bad cd's
The BUF-03 I thought with it's far low output impedance and far higher current would be even more of the above. But no it does exactly what pure class A poweramps do, things seem mellower/relaxed (almost tube like) it's all there but it's toned down a little like someone has slipped a tube in somewhere, but more so like a good class a/b poweramp which has been biased up almost into full class A.
Both are great to listen to but as to say which is right, I cannot say yet, I need to listen much longer to see if I miss the 844's attitude. One thing though the BUF-03 has got the thumbs up for driving the Sennheiser phones.

Cheers George

Hi George, You summed it up pretty well. I run my BUF03's with even larger heatsinks. I listen to Grado SR80's and Audio Technica ATH-M50's. 35 Ohm and 38 Ohm headphones. You can null those BUF03's out too. My current Preamp/Headphone amp is based on LME49710HA and BUF03, I use a 21 position attenuator at 20K Ohm. Compares well with my Nelsen Pass BA-3. The headphone amp is more interesting in the lower mid bass. I switch off the power amp first to avoid the thump. Also I noted that the AD version BUF03AJ sound a touch different then EJ's. Not sure why. PMI was the original design and AD bought them out sometime in the late 80's or early 90's. Lots of high end manufactures used the BUF03. Just a thought.... In my testing I find the BUF03 will take on a different sound based on what feeds it. If I stick it behind the BA-3 it sounds a bit like a OPA627, a tad dark like the DIP version. Behind the LME49710HA it isn't the slightest bit dark. Dave

 

[Torchwood421]

BUF03

Pic of the temp setup for those who wish to see.

Cheers George

Looks good. Surprised to see a purpose built PC board. I'm using larger electro's with Wima bypasses on mine. My soic AD844's should be here today.

 

[Torchwood421]

The BUF03 headamp/PRE

Here it is on a plexiglass sheet.

 

[dacen]

Have now double stacked the I/V section of the AD844. Georgehifi is absolutely right. This is the most impressive improvement I have done in quite some years. Fantastic! For those of you that are into using the AD844, I can only applaud to this solution. Well done!!!

 

[dacen]

I have settled for 3.6 K and 680pf Styroflex on the pin 5 (2xAD844). My DAC is a PCM63. This was what made it really sing to my old ears.

 

[georgehifi]

Looks good. Surprised to see a purpose built PC board. I'm using larger electro's with Wima bypasses on mine. My soic AD844's should be here today.

Yes this was aready made board I picked up on ebay, I also have .1uf bypass caps underneath.

Have now double stacked the I/V section of the AD844. Georgehifi is absolutely right. This is the most impressive improvement I have done in quite some years. Fantastic! For those of you that are into using the AD844, I can only applaud to this solution. Well done!!!

That's three now out of three that have bothered to try stacking the I/V section of the AD844 with the same great results that I achieved, maybe this will make the nay sayers think again before questioning stacking the AD844's I/V stage in open loop configuration without even trying it. Current starvation with dac's glitches or transients seems to be the issue here I feel which is addressed with stacking, but it could also be the fact the input impedance is halved for each doubling of 844's i/v stages as well, it's win win whatever is the cause for the massive improvement in sound quality.

Cheers George

 

[georgehifi]

I have settled for 3.6 K and 680pf Styroflex on the pin 5 (2xAD844). My DAC is a PCM63. This was what made it really sing to my old ears.

That's -3db at 65khz which is close to what I use now, try 560pf with your 3.6kohm this will give a -3db at 79khz.
I found the magic number to be -3db at 75khz, 4.7kohm TZ resistor with 470pf. Flat to 20khz, minimal phase shift in the audio band 10khz-20khz yet still filtering out the dac's rubish.

Not that it matters, your so close anyway, just what I observed on the scope and listen to.

Cheers George

 

[georgehifi]

Just thinking out loud this afternoon, I now have an amazing amout of different configurations I can use, all using the double stack I/V stage of the AD844, all using the same value TZ resistor and cap.
I can do all these.

Drive the rca outputs:
1: Straight from the TZ point (a little lame, but nice)
2: From the BUF-03 buffer, driven from TZ
3: Using the internal 844 buffer
4: Using the BUF-03, driven from the internal 844 buffer.

And all these will be dc coupled throughout from dac output to rca ouputs.

I'm using 2 at the moment but itching to give 4 a go soon.

Cheers George

 

[dacen]

I am sure this design also could benefit from the AD844 I/V stacking.

 

[georgehifi]

I am sure this design also could benefit from the AD844 I/V stacking.

It's been done dacen a few pages back, http://www.diyaudio.com/forums/digital-source/227677-using-ad844-i-v-35.html#post3476672
"Supra" Mick Meloney of Supratek had a Pedja based TDA1541, and he stacked 3 and said. Quote: "I've also tried the stacked 844's and can hear why George was so impressed-I've never heard a TDA1541 sound like this, and I've owned and built more than a few. I used 3 stacked 844's but ordered more as probably need to put another one on to suit the 1541.
As George says it is very dynamic, very defined, none of the "euphonic" sound commonly found with 1541's."

Cheers George

 

[dacen]

That´s great. I have one DAC with two TDA-1541 S1 single crown in diff. mode. I will certainly modify that one with the AD844 just to see how it sounds. This is promising!! Sorry I forgot SUPRA´s comment. I remember I have read that one earlier. I am just a bit enthusiastic right now. It is not every day I come across a modification with such an impact.

 

[supra]

It's been done dacen a few pages back, http://www.diyaudio.com/forums/digital-source/227677-using-ad844-i-v-35.html#post3476672
"Supra" Mick Meloney of Supratek had a Pedja based TDA1541, and he stacked 3 and said. Quote: "I've also tried the stacked 844's and can hear why George was so impressed-I've never heard a TDA1541 sound like this, and I've owned and built more than a few. I used 3 stacked 844's but ordered more as probably need to put another one on to suit the 1541.
As George says it is very dynamic, very defined, none of the "euphonic" sound commonly found with 1541's."

Cheers George

George
I'm playing around with the OPA860 and OPA861 now, you should maybe check them out. Rogic used these too, that guy was/is so far ahead of us.

 

[georgehifi]

George
I'm playing around with the OPA860 and OPA861 now, you should maybe check them out. Rogic used these too, that guy was/is so far ahead of us.

I did have a look at these, and have some samples somewhere, but the 844 for me was more versatile with it's built in buffer the TZ point and dc offset stable without feedback, and it takes + & -15vdc that's available in most cdp's and dacs, plus it's 8 pin dil not smd.
But I agree they are worth trying also, but I'm so pleased with the sound of the stacked 844's and still have not finished trying different things with it, now that I have the powerhouse BUF03 buffer to hang off it in two different ways. But at the moment I think the internal buffer of the AD844 still has the nod over the BUF03, but the 03 does have a certain tube niceness about it, all that Class A it's biased to I think, at the expense of of a bit of attitude I think, time will tell if the 844's buffer is still top dog.

Cheers George

 

[Torchwood421]

BUF03

I did have a look at these, and have some samples somewhere, but the 844 for me was more versatile with it's built in buffer the TZ point and dc offset stable without feedback, and it takes + & -15vdc that's available in most cdp's and dacs, plus it's 8 pin dil not smd.
But I agree they are worth trying also, but I'm so pleased with the sound of the stacked 844's and still have not finished trying different things with it, now that I have the powerhouse BUF03 buffer to hang off it in two different ways. But at the moment I think the internal buffer of the AD844 still has the nod over the BUF03, but the 03 does have a certain tube niceness about it, all that Class A it's biased to I think, at the expense of of a bit of attitude I think, time will tell if the 844's buffer is still top dog.

Cheers George

Hi George, It's all about the application. For driving low impedance the BUF03 is hard to beat. Keep a light load on the 844 and nirvana. All my playing around with chips shows it's all about tuning and trying things. Magic can happen. I'm down at the moment with an infection. I hope to get my 3 stacked 844's soldered up in a day or 2. Looking forward to taking a listen.

 

[skouliki]

I am listening the last couple of days to my AD1865 dac with stacked AD844(2x) for i/v conversion and i must admit it sounds really great. Maybe its a tiny bit less transparent than passive i/v( with opa627/buf634 gain stage) but the sound has huge dynamics and great bass. Im using on pin5 a 3k9 TX2352 resistor with 730pf styroflex capacitor and pin6 is output. Thanks georgehifi for sharing

Edit: Im using a Salas V1.0 to power the i/v stage at 12v, next step is to increase the voltage at 15v-16v

 

[georgehifi]

I am listening the last couple of days to my AD1865 dac with stacked AD844(2x) for i/v conversion and i must admit it sounds really great. Maybe its a tiny bit less transparent than passive i/v( with opa627/buf634 gain stage) but the sound has huge dynamics and great bass. Im using on pin5 a 3k9 TX2352 resistor with 730pf styroflex capacitor and pin6 is output. Thanks georgehifi for sharing

Edit: Im using a Salas V1.0 to power the i/v stage at 12v, next step is to increase the voltage at 15v-16v

That's 4 out of 4 now, with posistive praise for stacking.
That dac your using skouliki has almost the same output specs as my PCM1704 with +-1mA @ 1kohm. I have not got another 2 AD844 at the moment, but if you have 2 spare it would be very interesting to see what 3 stacked will do with this type of output, as 3 is definatly better than 2 with TDA1541.
Also the 1st order filter you are using is quite low at -3db @ 55khz. Give a 390pf a go with that 3k9k resistor. this then will give a first order filter of 104khz -3db. It may open up that transparency you mentioned.

Cheers George

 

[skouliki]

That's 4 out of 4 now, with posistive praise for stacking.
That dac your using skouliki has almost the same output specs as my PCM1704 with +-1mA @ 1kohm. I have not got another 2 AD844 at the moment, but if you have 2 spare it would be very interesting to see what 3 stacked will do with this type of output, as 3 is definatly better than 2 with TDA1541.
Also the 1st order filter you are using is quite low at -3db @ 55khz. Give a 390pf a go with that 3k9k resistor. this then will give a first order filter of 104khz -3db. It may open up that transparency you mentioned.

Cheers George

Good advice about the capacitor value. I will try to find some suitable styroflex. Actually i have one spare AD844 (bought them as five pack from ebay) - i could stack one channel with 3 AD844 temporarily and give it a listen. It should be interesting

 

[georgehifi]

Make that now 5 out of 5 for stacking the AD844's.
I just got word back from an analog lover customer, with Halcro DM68's, Lightspeed, Wilson Watt Puppy 8's. I did the same to his 303/200, as I did to my Cary 303/200, and he is over the moon on how good it sounds. He said he cannot believe that CD can sound this good, his words were, "it's like the best lp phono he's ever heard yet with all the good things that digital can give, separation, extension, dynamics, bass tightness power and control, and no noise penalty of lp"

Cheers George

 

[jcx]

increase the filter order, Butterworth is much flatter in the passband
should be inaudible at a much lower fc

since you need a buffer anyway why not make it a 2nd order LPF with Q, fc coordinated with the RC for a 3rd order Butterworth @ ~ 40-50 kHz