[aeriyn]

I ordered some different opamps a while back because I'd been thinking that the AD8620/AD8610 combo was too bright for my CD player. Because of my general laziness I never put them in until today. I wasn't too impressed with the AD845... it was too muddy.

But the AD843s... dayum. DAYUM. Everything I loved about the AD8610 and none of the bad sharpness... everything I loved about the OPA627, except with all the snap and "in your face" aspect of the Analog Devices opamps. The level of detail with the AD843s just stomps all over the AD86x0. I can hear BoA sucking in a breath between her words. I can hear the soft singing of backup singers I never even knew were in the song.

I'm sure that some of the improvement relative to the AD8620 is the fact that this is two single-channel opamps as opposed to a dual opamp, but probably more of it is how awesome this chip sounds.

I seriously hope I get a job soon... I am itching to build a better amp using AD843s.

I'm enjoying listening to it, and it's really a shame considering how I'm going to probably have to sell my pimeta and my cd player to pay rent soon...

 

[Syzygies]

In principle one could use AD843s in a MINT:

Single-to-Dual Op-Amp Adapter - SO8 to SO8 version

Sorry to hear about your liquidity crisis!

 

[aeriyn]

AD843s might not be so great for portable though; they draw more power than OPA627/637!

 

[Natsuiro]

I really like the 843's as well. I had them in my PPA before I sold it =)

 

[Tomo]

Hi,

I recall AD843 was running Class-A-ish. Which explains higher quiescent current. But then most of us bias our opamps to class-a anyways.

T

 

[aeriyn]

They can't run unbuffered as far as I can tell. They oscilliate like crazy in a cMoy type design; yes I've tried.

 

[Tomo]

Hi,

That is weird. AD843 is supposed to be Unit Gain Stable. Oscillation may come from the feedback line. Try using a damping capacitor in the feedback. (I think it a VFB, so no worries about it causing problems.)

T

 

[aeriyn]

Quote:

Originally Posted by Tomo Hi, That is weird. AD843 is supposed to be Unit Gain Stable. Oscillation may come from the feedback line. Try using a damping capacitor in the feedback. (I think it a VFB, so no worries about it causing problems.) T

It is unity gain stable, I'm successfully using one on the "ground" channel of my pimeta amp. But I cannot seem to get a cMoy type design to not oscilliate with this opamp, although it was only because I was bored that I even tried.

However, there was no problems of oscilliation or any other problems when replacing AD8620 with two AD843s in my pimeta's L and R channels... but perhaps my ugly DIY adapter made of perfboard and cat5e wire is not the best for hooking it up to the amp, though.

 

[Tomo]

Awww ...

I am sorry. I know a company who used to sell CMOY-like headphone amp (good olde negative feedback non-inverted) with AD843. So I think if you spend more time tinkering I am pretty sure you end up with good results.

This sorta thing is due to a lot of things. One of my amp had problems from not having shielded input cables, but this is not so with others. Another one just needed very good bypassing at the opamp's power supply pins, which not many people encountered here.

Of course, one of them reason can be a human error. I remember once I forgot to install signal coupling capacitors in Szekeres amp. Oh boy I had a massive DC offset going to the CDP AND the Headphones. I have no clue how both of them survived.

(This last entry is for me. This is one of many TOMO blunders. ... Yeah I had a habit of grasping at soldering iron for a few years too. NO KIDDING!)

T

 

[steinchen]

Quote:

Originally Posted by aeriyn However, there was no problems of oscilliation or any other problems when replacing AD8620 with two AD843s in my pimeta's L and R channels

I had a little distortion at high volume levels with the ad843, adding 2.2uF tantalum and 100nF ceramic caps to the rail bypass caps (like recommended in the datasheet) fixed that.

also make sure to pick the right jfet(s) for rail noise isolation on PPA, the ad843 draws 13mA quiscient current (opa627 only 7mA)

 

[tangent]

Quote:

Originally Posted by aeriyn I'm sure that some of the improvement relative to the AD8620 is the fact that this is two single-channel opamps as opposed to a dual opamp

No, the 13mA quiescent is where that tubey gooeyness comes from. The AD8610, by contrast, is a very efficient little chip, considering how good it sounds. Nothing's free, though.

One thing to be careful about it, though, is that when annoyed it has a tendency to lock up its output near one rail. Can you say "massive DC offset"? Yes, I thought you could.

Quote:

They can't run unbuffered as far as I can tell. They oscilliate like crazy in a cMoy type design;

No, they can be made to work. You'll notice that I review this op-amp on my site, which means it worked for me in a CMoy. That amp is on a professional PCB, though, and I used the predecessor to the Brown Dog dual DIP-8 op-amp adapter, so all circuit paths are short and tight.

 

[aeriyn]

Quote:

Originally Posted by tangent One thing to be careful about it, though, is that when annoyed it has a tendency to lock up its output near one rail. Can you say "massive DC offset"? Yes, I thought you could.

I... don't like the sound of that... >.>;

What annoys an AD843?

My amp is still running at <2mV offset, so apparently it's not annoyed right now.

Quote:

Originally Posted by tangent That amp is on a professional PCB, though, and I used the predecessor to the Brown Dog dual DIP-8 op-amp adapter, so all circuit paths are short and tight.

Maybe my ghettofied adapter is what's causing the issues in the cMoy type design that I tested them on; however I get no adverse affects when I plugged it into my pimeta.

I'm looking forward to building another, better amp using these opamps.

 

[myself, aka me]

Quote:

Originally Posted by aeriyn but perhaps my ugly DIY adapter made of perfboard and cat5e wire is not the best for hooking it up to the amp, though.

That pissed off my OPA637 combined with a lower than normal voltage. I used resistor clippings. I need something better, yes.

 

[Dreamslacker]

Quote:

Originally Posted by aeriyn I... don't like the sound of that... >.>; What annoys an AD843? My amp is still running at <2mV offset, so apparently it's not annoyed right now. Maybe my ghettofied adapter is what's causing the issues in the cMoy type design that I tested them on; however I get no adverse affects when I plugged it into my pimeta. I'm looking forward to building another, better amp using these opamps.

You'd realize that the Pimeta PCB has film caps and small electrolytics near the opamp power pins whereas a Cmoy doesn't?
You can get the AD843 to work in a Cmoy, just place a 0.1uF film and a 2.2uF tantalum cap beside the socket connected to the -ve and +ve pins. That should help the chip behave better.

 

[aeriyn]

Thanks. It works fine now. But that was only for kicks, the AD843s were purchased for future PPA build.

Although, I'm still using them in the pimeta until I have incoming monies again.