[pabbi1]

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif Al: with so many amps, why not just build a single attenuator box and then have output selector to control each of your amps? Then, just pull all the pots and steppers out and be done with it

Now, Marc, you know better. It is my G_d given right to suffer intimately with every cluster forked component on the planet, in the most non-sensical combinations possible.

Not a bad idea, if only more than one of my amps actually worked at the same time, your suggestion would be fairly reasonable. Oh, and Alex would be building me a LOT of cable.

Seriously, who woulda thunk attenuators that I have used in other applications would fail in such a way (assuming I replicated the same wiring diagram that I have executed in at least 2 previous build successfully) ?

Again, it brings us back to the shrinking possibilities of attenuatorsto use, the JT and it's ilk not withstanding.

The Goldpoints are excellent in the BH, and probably should stick with them, but there ain't nothing like the feel of something new...

 

[luvdunhill]

Quote:

Originally Posted by pabbi1 /img/forum/go_quote.gif Again, it brings us back to the shrinking possibilities of attenuatorsto use, the JT and it's ilk not withstanding.

well, if you want to stay off the beaten path... here are 2 options:

1.) optocoupler based attenuation, similar to what Melos (tried) to use. This is the relevant thread:

diyAudio Forums - Lightspeed Attenuator a new passive preamp

Best to start where Nelson Pass gets involved in the later pages and adds his 2 cents. A pretty straightforward project, assuming you buy enough black boxes from Allied and have time to match them. Bonus is you get a volume control that "feels" like a pot.

2.) transformer based attenuation (TVC / AVC). I'm exploring this option now, and have a custom unit coming my way for a project I'm finishing up. I think one of the better options is offered my Mr. Slagle:

intact audio

I ended up not using an autoformer, but a true TVC instead. We will see how it goes.

have fun!

 

[MrMajestic2]

Has the RK40 been discontinued? I have only seen it as a linear stereo pot on FamilyGate on ebay and hardly anywhere else.

 

[luvdunhill]

Al:

sorry got interrupted by the babies.. anyways, the third option I'd consider would be something like the PGA2310. You'll have to look around for boards, but they exist. Also, check out what Ebay seller AudioTek has for sale. Very reliable seller and one I'd highly recommend. Here are a few attenuators for sale there:

eBay Store - AudioTek Shop: Volume Control Attenuator: ALPS RK27 Attenuator VR Volume Control 100K Stereo

 

[Pars]

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif Al: sorry got interrupted by the babies.. anyways, the third option I'd consider would be something like the PGA2310. You'll have to look around for boards, but they exist. Also, check out what Ebay seller AudioTek has for sale. Very reliable seller and one I'd highly recommend. Here are a few attenuators for sale there: eBay Store - AudioTek Shop: Volume Control Attenuator: ALPS RK27 Attenuator VR Volume Control 100K Stereo

I had a Coda Window preamp for awhile last fall that used IIRC a PGA2310 in it. My impression regarding resolution of it was approx. equivalent to an Alps Blue. Lots of steps, etc. but ultimately not in the same domain as a TKD, stepped attenuators, RK50s, etc. Apparently Crystal Semi's competing dig. volume chip was better, as it did not integrate an opamp in it so that was not a limiting factor. I have been told they discontinued theirs however.

 

[00940]

wrt digital pots, Jocko suggested on diyhifi to have a look at the maxim DS1666.

 

[amb]

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif with so many amps, why not just build a single attenuator box and then have output selector to control each of your amps? Then, just pull all the pots and steppers out and be done with it

Actually this is not a good idea.

The wiring after the pot to the amp should be kept very short due to its high impedance nature. When you put the pot in a separate box, add a switch with a bunch more wiring, and then another interconnect cable to the amp, you subject the signals to capacitive coupling between channels and to ground (increased stereo crosstalk, high frequency rolloff) as well as EMI/RFI.

Same reason why a "passive preamp" without active output buffering is a bad idea.

 

[fzman]

Quote:

Originally Posted by Pars /img/forum/go_quote.gif I had a Coda Window preamp for awhile last fall that used IIRC a PGA2310 in it. My impression regarding resolution of it was approx. equivalent to an Alps Blue. Lots of steps, etc. but ultimately not in the same domain as a TKD, stepped attenuators, RK50s, etc. Apparently Crystal Semi's competing dig. volume chip was better, as it did not integrate an opamp in it so that was not a limiting factor. I have been told they discontinued theirs however.

not sure where you got the info that the crystal semi part was better, as coda upgraded to the pg2310 and found it an improvement. the chip did have progarmable gain, and according to coda, sounded better with some of the gain activated. i have a coda preamp that started its life with the crsyal part, and was upgraded to the pg part. it sounded better.

 

[BradJudy]

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif Al: sorry got interrupted by the babies.. anyways, the third option I'd consider would be something like the PGA2310. You'll have to look around for boards, but they exist.

I found this Danish place a while back that sells pga2310 kits (look for ChipVol1) - Dantimax - Remote control audio kits

 

[luvdunhill]

Quote:

Originally Posted by amb /img/forum/go_quote.gif Actually this is not a good idea. The wiring after the pot to the amp should be kept very short due to its high impedance nature. When you put the pot in a separate box, add a switch with a bunch more wiring, and then another interconnect cable to the amp, you subject the signals to capacitive coupling between channels and to ground (increased stereo crosstalk, high frequency rolloff) as well as EMI/RFI. Same reason why a "passive preamp" without active output buffering is a bad idea.

excellent point. what does the output buffering do to alleviate these issues with a passive preamp? Seems that after the signal leaves the buffers, the same issues would crop up.

 

[Pars]

Quote:

Originally Posted by fzman /img/forum/go_quote.gif not sure where you got the info that the crystal semi part was better, as coda upgraded to the pg2310 and found it an improvement. the chip did have progarmable gain, and according to coda, sounded better with some of the gain activated. i have a coda preamp that started its life with the crsyal part, and was upgraded to the pg part. it sounded better.

I believe there are/were discussions on diyaudio or diyhifi regarding these, and those that had used both seemed to prefer the Crystal parts. I have only heard what was in the Coda I had, so I guess I should not have commented with hearsay.

 

[cetoole]

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif excellent point. what does the output buffering do to alleviate these issues with a passive preamp? Seems that after the signal leaves the buffers, the same issues would crop up.

Nope, the buffer acts as an impedance converter, so you get a much lower output impedance, which means you need much more capacitance to get the same levels of capacitive coupling as a passive "preamp". Also, the low impedance signal is less susceptible to picking up interference. The buffer doesnt need to be anything complex, it could be something as simple as a pair of JFETs, like the new Pass B1. Of course, these particular issues only really affect a resistive attenuator, TVC or AVC should be relatively immune.

 

[amb]

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif excellent point. what does the output buffering do to alleviate these issues with a passive preamp? Seems that after the signal leaves the buffers, the same issues would crop up.

A buffer (having high input impedance and low output impedance, the latter being the more important factor in this discussion), located physically close to the pot, lends the signal a high level of immunity to capacitive coupling and interference issues.

 

[luvdunhill]

for some reason I was thinking the output impedance of the joshua tree was higher, but it looks like it's around 750 ohms... so, perhaps not such a bad idea after all?

oh, and for the person that recommended the DS1666, look at the DS1802.. it's used in some commercial gear that I think highly of, so it should be worth investigating more.

 

[amb]

Yeah, 750 ohms is not too high, but is it 750 ohms in any attenuator position? As far as I know the JT is completely passive (relay switches in/out a bunch of resistors to make a variable voltage divider), in order to have an output Z of 750 ohms or less in all positions, the resistors will have to have very low values, which means it could be a difficult load to drive for the preceding stage. I couldn't find a schematic on twistedpearaudio.com website's JT page, so I couldn't verify.