|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
63.246.183.22
I use SPICE whenever possible to analyze circuits I'm planning to construct. This evening I combined an inverse RIAA filter with the RCA phono preamp circuit. The result is a frequency response plot that's essentially flat over much of the audio spectrum, but there's a pronounced high end rolloff. At 20KHz, amplitude is down 3dB. I'm pretty sure this isn't related to networks in the preamp, so maybe it's the Miller effect at the 12AX7s. Anyone here ever used an inverse RIAA to measure their phono pre?
--------------------------
Buy Chinese. Bury freedom.
Follow Ups:
I've built this a few times, also using 6SL7 as JE Labs did (what ever happened to Joseph?). I completely agree a light load is needed to prevent rolloff. If you look at equipment of this era, not uncommon to see 1meg volume controls, of course there was often a 500k balance control too, meaning (without doing the math) something in the 250k range was not only important, but likely to already be there. Personally I don't like using sand (just my ethic) and I don't want to be limited by such a light load, so I use a 6922 as the second stage now (completely different circuit overall, including shunt reg). That avoids a cathode follower and allows a 50k to 100k input impedance.
RCA was not "kidding" about keeping the external load impedance up. The 'X7 triode is a wimp and truly "can't fight its way out of a wet paper bag".
Downstream drive was an item I addressed in the tweaked version of the setup.
Eli D.
Eli, the modified 6N2P version is your creation, am I right? If so, I'm wondering how you arrived at the modified component values. Specifically, I'm referring to the series cap and resistor at the anode of the first stage. Was this done to accommodate the 6N2P, or is it due to the change in loading (second stage now being grid leak)? More important, how did you confirm these values for accuracy, and why does each component show a range of values, rather than a single fixed value?
--------------------------
Buy Chinese. Bury freedom.
Edits: 07/22/17
The changes accommodate the grid leak biased 2nd gain block. I had help with the values, from another "inmate" who (unlike me) really knows his way around RIIA EQ networks.
IMO, the differences (if any) between 6H2П-EB (6n2p-ev) and 12AX7 are far too small to be important.
The grid leak bias idea comes from Thorsten Lösch and improves bass extension by lightly loading the RIAA EQ network.
Eli D.
Thanks, that's helpful to know. I don't see a problem with bass extension in the RCA design, so I'll probably stick with the original values and topology. I do like the improvement with the 24k resistor that you noted. It's clear too me now that I'll need to build an inverse RIAA to verify accuracy, regardless of the final design.
--------------------------
Buy Chinese. Bury freedom.
The rolloff is there even with no load. I spent a little more time with it, and the culprit is the 180pF across the output. That's supposed to represent the capacitance of the output cable. In my case, there will only be a few inches between this circuit and the CF that follows it. Oh well, this exercise gave me an reason to sharpen the accuracy of the inverse RIAA that I'm using for the simulations. Everything looks good now.
--------------------------
Buy Chinese. Bury freedom.
Edits: 07/20/17
I don't generally use inverse RIAA, but I have plotted a graph with an RIAA curve and my actual measurements from said phono preamp.
I think what you're seeing is indeed related to Miller capacitance. I wouldn't ever drive a 12AX7 with another 12AX7 unless I was building a guitar amp.
Thanks, for this graph work. Wonder what the FR would be for a 5751? In terms of HF roll-off.
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: