30 VA is still small and convenient, and not that much more expensive. I prefer to overspec. a little (or a lot depending on your viewpoint).
You can use whatever you like. There's quite a lot of older discussion on this topic buried in the thread that you can access with the search function.
You can use whatever you like. There's quite a lot of older discussion on this topic buried in the thread that you can access with the search function.
the test for a power supply is to place a capacitor from the rail to the input of the RIAA eq amp and test for hum and noise. the cap should be no larger then .1 uf.
in mine, there is NO audible difference between B+ , B- and ground.
I kept the HA1457 in the sx 780 with new RIAA curve as the performance is excellent and the rail voltage is -+ 23 volts.
The new RIAA is to match the 26db flat head amp installed into the system.
the high frequency gain of the HA1457 is 7 instead of 34.
in mine, there is NO audible difference between B+ , B- and ground.
I kept the HA1457 in the sx 780 with new RIAA curve as the performance is excellent and the rail voltage is -+ 23 volts.
The new RIAA is to match the 26db flat head amp installed into the system.
the high frequency gain of the HA1457 is 7 instead of 34.
Last edited:
"a" test for a power supply.
I mean sure, whatever works for you I guess though if you think about it the procedure makes little sense. At least not in the present time when anyone can measure the output noise spectrum precisely with a soundcard and free software like RMAA. Amplifying the voltage rail AC component through a high gain preamplifier doesn't give you any additional useful information.
I mean sure, whatever works for you I guess though if you think about it the procedure makes little sense. At least not in the present time when anyone can measure the output noise spectrum precisely with a soundcard and free software like RMAA. Amplifying the voltage rail AC component through a high gain preamplifier doesn't give you any additional useful information.
No value because it doesn't factor in the PSRR of the amplifier, and meanwhile amplifies the supply rail noise by a variable amount depending on the gain of the amplifier.
The test can confirm an unspecified amount of noise on the V+ rail. It doesn't tell you if it matters or not.
The best way to answer that question is to simply measure the output noise spectrum under normal operation. Is power supply noise visible above the baseline output noise of the amplifier Y/N? If TRUE then fix if noise level is considered undesirable*. If FALSE then done.
*Lots of circuits have residual power supply noise and ripple in the output, power amplifiers in particular. For a high gain, high PSRR, relatively low S/N circuit like a solid state phono preamplifier however it's not too onerous to remove it completely and generally desirable to do so.
The test can confirm an unspecified amount of noise on the V+ rail. It doesn't tell you if it matters or not.
The best way to answer that question is to simply measure the output noise spectrum under normal operation. Is power supply noise visible above the baseline output noise of the amplifier Y/N? If TRUE then fix if noise level is considered undesirable*. If FALSE then done.
*Lots of circuits have residual power supply noise and ripple in the output, power amplifiers in particular. For a high gain, high PSRR, relatively low S/N circuit like a solid state phono preamplifier however it's not too onerous to remove it completely and generally desirable to do so.
Last edited:
...
Without testing anything you can work out the requirements by hand easily enough:
e.g. An OPA134 has 90 dB PSRR at 120 Hz, and 8 nV/sqrtHz input referred noise.
The maximum allowed 120 Hz ripple on the power supply is therefore 8 nV x 30000 x 2.8 = ~1 mV p-p.
Such V+ ripple fed directly to the OPA134 input would most likely lead to audible hum. When the V+ is connected to the power pins, however, the PSRR mops it all up and the output signal is clean. There is no advantage to reducing the power supply ripple below 1 mV as no further reduction in output noise is possible.
Without testing anything you can work out the requirements by hand easily enough:
e.g. An OPA134 has 90 dB PSRR at 120 Hz, and 8 nV/sqrtHz input referred noise.
The maximum allowed 120 Hz ripple on the power supply is therefore 8 nV x 30000 x 2.8 = ~1 mV p-p.
Such V+ ripple fed directly to the OPA134 input would most likely lead to audible hum. When the V+ is connected to the power pins, however, the PSRR mops it all up and the output signal is clean. There is no advantage to reducing the power supply ripple below 1 mV as no further reduction in output noise is possible.
Last edited:
Yes, the lower the PSRR the more effort you have to spend to reduce noise/ripple on the V+.
A simple Jfet amplifier stage with resistive (non CSS) load has ~0 dB PSRR making these one of the hardest circuit elements to deal with as a low noise preamp. I tried it. My conclusion at the end of the day was it wasn't worth the effort: the simplicity gained with the low parts count JFET circuit was completely countered by all the hoops needed to jump through on the power supply side. I mean, it worked fine but it just seemed to me I was playing a zero sum game. YMMV
A simple Jfet amplifier stage with resistive (non CSS) load has ~0 dB PSRR making these one of the hardest circuit elements to deal with as a low noise preamp. I tried it. My conclusion at the end of the day was it wasn't worth the effort: the simplicity gained with the low parts count JFET circuit was completely countered by all the hoops needed to jump through on the power supply side. I mean, it worked fine but it just seemed to me I was playing a zero sum game. YMMV
The datasheet notes it substitutes for the BB INA217 so you can consider that as an alternative.
You aren't the first (or the last) person to want to use instrumentation amplifiers for phono preamps. You can probably find someone more capable than I to explain the pros and cons of it, but my own feeling is that the common mode noise in the cartridge output signal is simply not significant enough to be worth the -6 dB noise penalty associated with running a balanced input.
(each side only sees half the cartridge signal but both halves of the INA have about the same input-referred noise as any regular low noise op amp, so the S/N ratio is effectively halved)
You aren't the first (or the last) person to want to use instrumentation amplifiers for phono preamps. You can probably find someone more capable than I to explain the pros and cons of it, but my own feeling is that the common mode noise in the cartridge output signal is simply not significant enough to be worth the -6 dB noise penalty associated with running a balanced input.
(each side only sees half the cartridge signal but both halves of the INA have about the same input-referred noise as any regular low noise op amp, so the S/N ratio is effectively halved)
Last edited:
An externally hosted image should be here but it was not working when we last tested it.
Waiting for coupling caps(russian nos) Not using smoothing caps on board as im using a super regulator(ssr03). Not planning on upgrading to MC anytime soon, so im using various loading caps on the switch board. 33, 100, 100, 150[pF].
Should have turned the dale resistors the other way with value up, but darn it looks so cute now
Missing a pair of jumpers on pics.
Hello how are you. Today I annoy them because I need help.
I always play with simulations in LTSPICE. I usually have simulations of phono preamps.
I need you to help me "copy and paste" a Laplace font for the inverted RIAA function.
I'm amateur and I can not do it.
Thanks friends of the forum.
I always play with simulations in LTSPICE. I usually have simulations of phono preamps.
I need you to help me "copy and paste" a Laplace font for the inverted RIAA function.
I'm amateur and I can not do it.
Thanks friends of the forum.
You can cheat by setting up a little passive filter circuit to generate the inverse RIAA function. It's a simulation so you can make the output impedance as low as you want i.e. use super-low resistance values and ridiculously large capacitance.
I'd share but I don't actually use this myself. I take the output of the circuit using a flat voltage signal source and compare it in an Excel worksheet later with the reference RIAA curve. I've attached an example.
If anyone does know how to generate the inverse RIAA function directly in LTSpice using the transfer function, please feel free to jump in here - I'd also like to know how to do it.
I'd share but I don't actually use this myself. I take the output of the circuit using a flat voltage signal source and compare it in an Excel worksheet later with the reference RIAA curve. I've attached an example.
If anyone does know how to generate the inverse RIAA function directly in LTSpice using the transfer function, please feel free to jump in here - I'd also like to know how to do it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
The formula of reverse RIAA with trial and error arose.
Where in RIAA was multiplied I put it divided and etc. (The formula of Laplace RIAA was given to me by Calvin)
I think it was fine, but I submit it to corrections.
Do not forget that I am a doctor, not an engineer.
Last edited: