|13th March 2020, 02:43 AM||#11|
Join Date: May 2018
Location: Cambridge UK
If you use a bipolar opamp in that tone section then the base pot may give issues as you are passing the opamp bias current through its wiper.
|14th March 2020, 02:23 AM||#12|
Join Date: Nov 2004
Besides, I would use this relatively noisy tone circuit at line level only, with the volume pot following. I would use maybe a 20k volume pot and another opamp to provide anywhere from unity gain to +6 dB.
Actually the 5532 should work well for that position then, levels at this point are low (so common-mode distortion should be a non-issue) and you would have no problem getting feedback network impedance low enough, e.g. 1k/1k, as the part is a good load driver. (It would need an input capacitor of maybe 10 µF and a ca. 47 kOhm bias resistor to ground though; I would also add 220-330 ohms in series with the noninverting input for stability reasons.)
I found another version of the sample circuit, the disappearing resistor is a 330 ohm there, so Av = 40 dB. Output noise is given as 1.2 mV, which is rather unacceptably high in my book. (The STK4xxx2II series has the same spec, it's a maximum there. Not very meaningful.) I did not find a spec for minimum stable gain, but knowing these old circuits, it may not be lower than 30-34 dB.
FWIW, I turned up an amplifier using the larger STK-461 (2x20 W), the Akai AM-U01, with schematic and specs. It uses almost exactly the application circuit and is given with an SNR spec of 90 dB. That would be ca. 400 µV of output noise, rather more in line with a (run of the mill) 1980 vintage hi-fi amplifier and indicating ca. 4 µV worth of effective input noise.
You know what? I would want to use a bit of a trick to bring gain on the STK down:
I think we can live with a boost of +0.2 dB at 10 kHz and +0.7 dB at 20 kHz.
Then the preceding opamp (5532) could be set up for a voltage gain of ca. +6.6 dB (for example, 1.2k + 1k) to make up for the lost gain. Most of it, anyway - a 15 wpc amp does not really need more than 37 dB of gain even for a 150 mV input sensitivity.
All of this would drop output noise from ca. 400 µV to 150 µV, far more in line with a decent amplifier. (The bottleneck is in fact still the STK...)
Note: The 6 MHz GBW assumed in simulation for the amplifier model is likely to be overly conservative. It is likely to be anywhere from 10 to 100 MHz in reality, possibly 20-30. An LM1875 (typical gain +26 dB) is spec'd with 5.5 MHz already.
Last edited by sgrossklass; 14th March 2020 at 02:38 AM.
|Thread Tools||Search this Thread|
|Thread||Thread Starter||Forum||Replies||Last Post|
|EXDF mode and how ak449X DAC's interpolator handle a hi res input?||laserscrape||Digital Line Level||0||14th October 2019 08:54 AM|
|PCB Routing Advice Needed: How to Handle Input/GND||butizzle||Construction Tips||7||6th April 2015 04:03 AM|
|STK459 replacement||Lumbermatt||Chip Amps||21||29th January 2015 08:07 PM|
|How to use STK459 in bridge mode||tridibesh||Solid State||0||4th June 2005 07:40 AM|
|Stk459 / Stk463||nbcd||Chip Amps||4||15th September 2004 11:51 AM|
|New To Site?||Need Help?|