I put together a passive pre this evening, based on the Harry Haller design, as modified and depicted in the schematic by planet10 (Dave) as shown in the original thread in March, 2002.
I built just one channel at first by paralleling two Caddock 22k's between input and output signal on the RCA's; a 22 ga single strand copper wire (one strand from a Cat5e cable) was run from output signal to the wiper of a Dale 100k audio taper pot, with the other appropriate end of the pot connected with another single 22 ga copper wire to both input and output grounds on the RCA's. I checked it out with my CD player (30 ohm output impedance per Cambridge Audio) connected through the passive to the power inputs of my NAD integrated (don't know that input impedance) and to the inputs of an Acurus A150 power amp (input impedance 20k ohms.)
Volume control was WAY too sensitive with either power amp--I couldn't run above 9 o'clock, so I built the other channel with a single Caddock 22k between input and output signal--and that got me to only about 10 o'clock. Still way too sensitive........
(Side note: Cables are all Belden 1505F terminated with Canare RCAP's; the CD to passive cables are 3 feet long; I tried both 2 foot and 4 foot cables from the passive to the power amps and it seems as if there are no issues with those cable lengths--I noted no attenuation of high frequencies and it certainly seems as if the CD player is quite comfortable with driving that length of cable through the passive.)
What should I do next? Do I try running two 22k resistors in series between input and output signal, or do I try one 22k between in and out signals with one 22k between output signal and ground (if I interpret another modification to the design from the original thread properly)?
Seems like sushi from Norway has the same problem--post #170 in the original thread states he built it just like me, but there were no replies to his question.
Thanks for any advice!
I built just one channel at first by paralleling two Caddock 22k's between input and output signal on the RCA's; a 22 ga single strand copper wire (one strand from a Cat5e cable) was run from output signal to the wiper of a Dale 100k audio taper pot, with the other appropriate end of the pot connected with another single 22 ga copper wire to both input and output grounds on the RCA's. I checked it out with my CD player (30 ohm output impedance per Cambridge Audio) connected through the passive to the power inputs of my NAD integrated (don't know that input impedance) and to the inputs of an Acurus A150 power amp (input impedance 20k ohms.)
Volume control was WAY too sensitive with either power amp--I couldn't run above 9 o'clock, so I built the other channel with a single Caddock 22k between input and output signal--and that got me to only about 10 o'clock. Still way too sensitive........
(Side note: Cables are all Belden 1505F terminated with Canare RCAP's; the CD to passive cables are 3 feet long; I tried both 2 foot and 4 foot cables from the passive to the power amps and it seems as if there are no issues with those cable lengths--I noted no attenuation of high frequencies and it certainly seems as if the CD player is quite comfortable with driving that length of cable through the passive.)
What should I do next? Do I try running two 22k resistors in series between input and output signal, or do I try one 22k between in and out signals with one 22k between output signal and ground (if I interpret another modification to the design from the original thread properly)?
Seems like sushi from Norway has the same problem--post #170 in the original thread states he built it just like me, but there were no replies to his question.
Thanks for any advice!
Am new to forum, please forgive any violations of custom.
If I understand your problem you have 11K (two 22K) in series with signal and a 100K pot to ground. 6 dB (not really very much) of attenuation will happen when pot is set to 11 K which is 11% of rotation for a linear pot. This means all meaningful attenuations happen in first few degrees of pot travel. Change to a smaller pot and/or a larger series resistor. 11K isn't bad I'd change the pot.
If you're going to all this trouble why use a pot? Pots are assumed to be somewhat evil even if very good ones are used. A quality switch and a set of fixed resistors will probably do a better job.
I just built a passive with 1 dB steps in the normal listening range. I now feel 1 dB is overkill, 1 1/2 to 2 dB is fine.
Mine has the following atenuation in dB: infinite, 47, 40, 34, 29, 25, 22, 20, 18, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 4, 2, and 0. If interested, see attached schematic. I used Vishay and Holco resitors - big $$$. Leave out steps to save money.
These values are fine for my amplifier but only the bottom 6 steps are useful at my friends house so check your attenuation at your normal listening level before you design and buy parts.
If I understand your problem you have 11K (two 22K) in series with signal and a 100K pot to ground. 6 dB (not really very much) of attenuation will happen when pot is set to 11 K which is 11% of rotation for a linear pot. This means all meaningful attenuations happen in first few degrees of pot travel. Change to a smaller pot and/or a larger series resistor. 11K isn't bad I'd change the pot.
If you're going to all this trouble why use a pot? Pots are assumed to be somewhat evil even if very good ones are used. A quality switch and a set of fixed resistors will probably do a better job.
I just built a passive with 1 dB steps in the normal listening range. I now feel 1 dB is overkill, 1 1/2 to 2 dB is fine.
Mine has the following atenuation in dB: infinite, 47, 40, 34, 29, 25, 22, 20, 18, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 4, 2, and 0. If interested, see attached schematic. I used Vishay and Holco resitors - big $$$. Leave out steps to save money.
These values are fine for my amplifier but only the bottom 6 steps are useful at my friends house so check your attenuation at your normal listening level before you design and buy parts.
Attachments
Thanks for the reply.......
I wanted to do this cheaply and just as a test of whether this design was as good as claimed--hopefully to bypass my a/v receiver in the HT system when all I wanted to do was listen to 2 ch music.
I ordered single Dale conductive plastic 100k log pots , and what I got are pots marked as 100k audio/log taper. However,they are actually linear--my VOM shows 0 at full CCW rotation, 46-47k at 12 o'clock, and 95-96k at full CW rotation........there's no way that the resistance is about 10k at 12 o'clock, as I understand it should be for a log pot. Vendor agrees with my assessment and is replacing them.
I'm using pots as shunts to save money vs. a stepped attenutator--please understand that I'm an engineer, but not an EE, which means I can understand why a circuit works but maybe only "sort of"..........what I took from the discussion on this design is that because the only thing in series in the signal path between in and out is the high-quality resistor(s), the quality of the pot that shunts the "excess" signal to ground would have a minor effect on quality. But I'm a chemical engineer who only got thru DC circuits in physics, no EE or AC stuff--so maybe that's not right...........please feel free to correct me if I'm wrong!
Thanks again!
I wanted to do this cheaply and just as a test of whether this design was as good as claimed--hopefully to bypass my a/v receiver in the HT system when all I wanted to do was listen to 2 ch music.
I ordered single Dale conductive plastic 100k log pots , and what I got are pots marked as 100k audio/log taper. However,they are actually linear--my VOM shows 0 at full CCW rotation, 46-47k at 12 o'clock, and 95-96k at full CW rotation........there's no way that the resistance is about 10k at 12 o'clock, as I understand it should be for a log pot. Vendor agrees with my assessment and is replacing them.
I'm using pots as shunts to save money vs. a stepped attenutator--please understand that I'm an engineer, but not an EE, which means I can understand why a circuit works but maybe only "sort of"..........what I took from the discussion on this design is that because the only thing in series in the signal path between in and out is the high-quality resistor(s), the quality of the pot that shunts the "excess" signal to ground would have a minor effect on quality. But I'm a chemical engineer who only got thru DC circuits in physics, no EE or AC stuff--so maybe that's not right...........please feel free to correct me if I'm wrong!
Thanks again!
Unfortunately, the 'signal path' idea you describe above is a common misconception. The pot still needs to be of high quality because if it is not then the amount of current shunted away is 'unstable' (or whatever un-audiophile term you wish to use) thus the signal reaching the destination is also 'unstable'.
There is a small advantage to this particular design, in that less of the pot track is used as it is swept lower, so if the pot is bad, the impact is reduced with lower pot settings.
Thanks for clearing that up..........I didn't totally buy the idea that the quality of the shunt pot was not important, but I didn't have the knowledge to refute the notion, either!
Some folks have described the quality of the Dale pots that will be used in the final version as better than Alps Blue Velvet, which, as far as pots go, seem to be respectable (for lack of a better term.)
I can state the following, with no reservations--with the passive between my CD player and the power section in my NAD C320BEE or into the Acurus A150, rather than using the pre section of the NAD into either power amp, there was an obvious increase in transparency and reduction of grain in the mids and highs. I thought the NAD was pretty good--especially when compared to my a/v receiver in "pure audio" mode (supposedly bypassing most of the DSP functions) when driving the Acurus on the front channels. I've not made up my mind about the low end yet, but my initial impression was any potential deficiencies there were more than compensated by improvements higher up.
Once I get the right pots in the thing, I'm sure it will be the best $50 I ever spent on audio stuff (including the case, Cardas RCA's, shipping, etc.)
Some folks have described the quality of the Dale pots that will be used in the final version as better than Alps Blue Velvet, which, as far as pots go, seem to be respectable (for lack of a better term.)
I can state the following, with no reservations--with the passive between my CD player and the power section in my NAD C320BEE or into the Acurus A150, rather than using the pre section of the NAD into either power amp, there was an obvious increase in transparency and reduction of grain in the mids and highs. I thought the NAD was pretty good--especially when compared to my a/v receiver in "pure audio" mode (supposedly bypassing most of the DSP functions) when driving the Acurus on the front channels. I've not made up my mind about the low end yet, but my initial impression was any potential deficiencies there were more than compensated by improvements higher up.
Once I get the right pots in the thing, I'm sure it will be the best $50 I ever spent on audio stuff (including the case, Cardas RCA's, shipping, etc.)
I am having the same problem. I am using a 50k pot and the volume is either ON or OFF. I was looking at the Harry Haller passive pre which puts a 22k resistor in series with the 100K pot and an 18k resistor from the input to the ground. This looks like a solution but I think I remember peter daniel saying that you dont want resistors in series before the pot or it will degrade sound.
sejarzo
how has your solution worked out for you?
Thanks for the help.
Milo
sejarzo
how has your solution worked out for you?
Thanks for the help.
Milo
I'm guessing Milo has the same problem I did--a pot marked with an "A" in a spot such that someone along the way mistakenly presumed that it was an audio taper pot, but it's not.
With the replacement 100k audio taper pots (verified as such with a VOM), volume control is quite a bit better. All in all, though, with 11k between input and output signal and 0-100k from output signal to ground, attenuation at "full volume" (anywhere between about 1 o'clock and max CW rotation) is still a bit too much, and control remains a bit imprecise at low-to-mid volume levels.
I'm still amazed at how good it sounds, though--imaging is decidely more natural (more depth, much less unnatural forward projection of vocals) and the high end glare is gone.
I'll have to educate myself a bit more before constructing version 2, I guess!
With the replacement 100k audio taper pots (verified as such with a VOM), volume control is quite a bit better. All in all, though, with 11k between input and output signal and 0-100k from output signal to ground, attenuation at "full volume" (anywhere between about 1 o'clock and max CW rotation) is still a bit too much, and control remains a bit imprecise at low-to-mid volume levels.
I'm still amazed at how good it sounds, though--imaging is decidely more natural (more depth, much less unnatural forward projection of vocals) and the high end glare is gone.
I'll have to educate myself a bit more before constructing version 2, I guess!
Well it was panasonic pot from digikey. I didnt know at the time that 50K was totally overblown.
Because the only 10k pot that I could find a a reasonable price was a liner taper I went ahead and got it for the replacement and I plan to set it up like ESP's "false log" pot. Anyone think I'm nuts, or have a better source for 10K pots?
Thanks guys.
Milo
Because the only 10k pot that I could find a a reasonable price was a liner taper I went ahead and got it for the replacement and I plan to set it up like ESP's "false log" pot. Anyone think I'm nuts, or have a better source for 10K pots?
Thanks guys.
Milo
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.