I was experimenting with running different preamps into a Hypex NCore amplifier. I modified the Hypex buffer to create a unity gain buffer so I have the right gain structure, by removing Rg shown here. Works great!
One preamp I have is singled ended tube, AC coupled through a cap, Output Z of 1500 Ohms. If I measure the frequency response into 100K ohms, I get a LF rolloff of almost 5 db (I should increase the cap size - I think its 1or 2 uF).
When I run this preamp into the Hypex, and measure at the output of the preamp + amp, the rolloff is completely gone. It's now flat to 20 Hz. Great!!
The Hypex has has an input impedance of 47K, but yet it behaves like a very light load. THD is even down as well. I tried reading it about it, and I see that the feedback of the instrumentation amplifier buffer somehow simulates a high impedance. But I don't quite get it. Could someone explain how this works? And also, if it doesn't behave like a 47K load, why is the input R listed that way? To everyone trying to bypass buffers of amps, more transparency blah blah, just take a look at the benefits! See pic below.
One preamp I have is singled ended tube, AC coupled through a cap, Output Z of 1500 Ohms. If I measure the frequency response into 100K ohms, I get a LF rolloff of almost 5 db (I should increase the cap size - I think its 1or 2 uF).
When I run this preamp into the Hypex, and measure at the output of the preamp + amp, the rolloff is completely gone. It's now flat to 20 Hz. Great!!
The Hypex has has an input impedance of 47K, but yet it behaves like a very light load. THD is even down as well. I tried reading it about it, and I see that the feedback of the instrumentation amplifier buffer somehow simulates a high impedance. But I don't quite get it. Could someone explain how this works? And also, if it doesn't behave like a 47K load, why is the input R listed that way? To everyone trying to bypass buffers of amps, more transparency blah blah, just take a look at the benefits! See pic below.
The output impedance of 1500R and a 1uF cap gives a flat response down to 10Hz for both 100K and 47K loads. The logical conclusion is that your test load is not what you think it is.
Could be.
To test the preamp I go RCA to an RCA to bnc adapter into a quantasylum qa400 here:
https://forums.melaudia.net/attachment.php?aid=11905
To test the amp out, I created a voltage divider with a 5.6 and 1R power resistor and then put a 1x scope probe at that junction (analyzer can only do 1V in). Other side of scope probe is bnc right to qa400.
From the manual:
“
The inputs have a series 68 ohm resistor, followed by a low-capacitance ESD protection device, followed by a shunt resistor of 100K ohms, followed by a series C of 10 uF. That signal is then presented directly to a JFET op-amp non-inverting input.”
To test the preamp I go RCA to an RCA to bnc adapter into a quantasylum qa400 here:
https://forums.melaudia.net/attachment.php?aid=11905
To test the amp out, I created a voltage divider with a 5.6 and 1R power resistor and then put a 1x scope probe at that junction (analyzer can only do 1V in). Other side of scope probe is bnc right to qa400.
From the manual:
“
The inputs have a series 68 ohm resistor, followed by a low-capacitance ESD protection device, followed by a shunt resistor of 100K ohms, followed by a series C of 10 uF. That signal is then presented directly to a JFET op-amp non-inverting input.”
What tube is your preamp pushing the signal with? 12AX7? Tungsol datasheet says plate resistance is 80000 ohms 100 v plate and 62500 ohms 250 v plate. That is way below 1500 ohms you state and higher than 47k quoted for your amp. Just the capacitance of a 2 m RCA coax cable stresses a dynakit PAS2 preamp with 12AX7, which I own. Coax capacitance is about 330 pf/m . 3 m of that alone can roll off highs. dynakit made sure their ST120 transistor power amp had 100kohm input impedance to be compatible with PAS2/3 preamps.
Since the PAS2 doesn't sound any better than an op amp mixer and draws 100 watts, I parked it in the attic. My tube ST70 with original 7199 board (1% HD) sounds worse than my solid state amps with .1% HD. 12AX7 7199 & 6CA7 tubes were great for 1970 when RCA 40636 transistors were $10 each and sounded bad due to 400 hz Ft. Time marches on.
Since the PAS2 doesn't sound any better than an op amp mixer and draws 100 watts, I parked it in the attic. My tube ST70 with original 7199 board (1% HD) sounds worse than my solid state amps with .1% HD. 12AX7 7199 & 6CA7 tubes were great for 1970 when RCA 40636 transistors were $10 each and sounded bad due to 400 hz Ft. Time marches on.
Last edited: