Hans,
I have to thank you once again for the great help you have offered, and all the time and effort you have put in to helping me.
This whole thread has been a great learning experience to me, and contains information and considerations that I haven't found anywhere else.
I have to thank you once again for the great help you have offered, and all the time and effort you have put in to helping me.
This whole thread has been a great learning experience to me, and contains information and considerations that I haven't found anywhere else.
Since you reacted so positive, I send you some more info regarding the effect of the 47K Ohm input resistance that is normally advised as input resistance.
However the cable capacitance from cart to phono-preamp is just as important. I have used for this example a cart with 600 Ohm/500mH, but for every MM element this may differ.
First graph, showing the deviation from the ideal RIAA curve versus frequency, is taken with a 47K input resistance and 3 different cable capacitances of resp 100pF (green), 150pF (blue) and 200pF (red).
The flattest result is with 150pF, but there is -3dB deviation from RIAA at 20kHz !
Second graph shows the situation with an input resistance of 80K and a cable capacitance of resp 30pF, 50pF and 70pF.
Not that 80K/50pF gives a superior response over 47K/150pF, being flat to 20kHz. However a cable of 50pF could be a bit short.
Alternatively, the phono preamp could be mounted in the turntable, with a real capacitor over the input of 50pF.
So be aware that both values, input resistance and (cable) capacitance are playing an important role, and not just the input resistance and that each input resistance needs a different capacitor value to match.
Hans
The best place for any low level signal amplifier is at the source of that signal.
Placing the RIAA pre-amp at the very closest to the cables coming out of the TT arm is the best location.
Not at the far end of the interconnects to some audio equipment rack.
Placing the RIAA pre-amp at the very closest to the cables coming out of the TT arm is the best location.
Not at the far end of the interconnects to some audio equipment rack.
Hello, I've been reading this thread and see the great knowledge and experience that is bringing about.
Now, from my ignorance because I have the basic knowledge that I have acquired over 36 years assembling audio circuits, I ask, it is so important if a Phono preamplifier is 80 dB or SR 120 db?
If I put the amplifier volume to maximum and approaching the ear to hear the speaker just a slight sssss and normal listening level it seems that nothing is connected, that is no longer enough ?.
Com Muffsy that happens, if it is connected and there is no music, there is no audible noise.
Some of you have come from simulators and has assembled the kit?.
The sound delivery is really spectacular for its price.
It has overcome a few that I possess. Of course I have not compared with high-end preamps, as my pocket will not let me, but I am convinced that "quire more expensive if not better."
I saw the other are those gárficas showing different responses to different strengths and capacities.
I think you omitted the effect of the mechanical parts of the cartridge, which may influence the frequency responses.
And finally, against the advice of placing phono preamplifier within the turntable, I think it's a good idea, but I see that it is advisable to add 50pF as cargo.
I measured the wiring turntable arm and throws between 45 and 50 pF.
Use Audio Technica AT 440 Mla. 3200 ohms and has 490mH and manufacturer recommended load 47k and 100 to 200 pF. My cable from the cartridge to the RCA has 125 pF so I did not add capacitance.
Greetings and I continue to learn with your contributions. Jose
Now, from my ignorance because I have the basic knowledge that I have acquired over 36 years assembling audio circuits, I ask, it is so important if a Phono preamplifier is 80 dB or SR 120 db?
If I put the amplifier volume to maximum and approaching the ear to hear the speaker just a slight sssss and normal listening level it seems that nothing is connected, that is no longer enough ?.
Com Muffsy that happens, if it is connected and there is no music, there is no audible noise.
Some of you have come from simulators and has assembled the kit?.
The sound delivery is really spectacular for its price.
It has overcome a few that I possess. Of course I have not compared with high-end preamps, as my pocket will not let me, but I am convinced that "quire more expensive if not better."
I saw the other are those gárficas showing different responses to different strengths and capacities.
I think you omitted the effect of the mechanical parts of the cartridge, which may influence the frequency responses.
And finally, against the advice of placing phono preamplifier within the turntable, I think it's a good idea, but I see that it is advisable to add 50pF as cargo.
I measured the wiring turntable arm and throws between 45 and 50 pF.
Use Audio Technica AT 440 Mla. 3200 ohms and has 490mH and manufacturer recommended load 47k and 100 to 200 pF. My cable from the cartridge to the RCA has 125 pF so I did not add capacitance.
Greetings and I continue to learn with your contributions. Jose
Hi Jose,
A lot of useful remarks.
To start with, a SNR of 65dB is enough to have absolute silence with the cart lifted from the LP, a figure that is easy to meet with current hardware.
Having 75dB gives some margin for future carts having less output, more than 75dB does not bring any additional benefit.
You are right that I have omitted the mechanical resonance from the cart.
This depends largely on the tip mass and damping of the suspension and differs quite a bit between carts.
However, mechanical resonance is quite counterproductive in keeping the phase intact at higher frequencies.
That's why some apply more damping with a lower input resistance and restore the frequency curve with an inverse first order filter low pass filter, resulting in a not very universal one-off.
Better is to use a cart with a low tip mass and a small resonance peak, avoiding all this trouble.
You can change input resistance and input capacitance when measuring with a test LP, and keep the phono-preamp universal.
That's where some background may help to understand the effect of these changes.
Hans
Hi Hans. Thanks for the answer, which has been very clear for someone who does not have enough theoretical basis like me (actually I'm a doctor, I hobbyist Audio and DIY. Today I have day off work so I am dedicated forums).
I will follow the thread as it is very interesting.
regards
I will follow the thread as it is very interesting.
regards
Thanks for the great info Hans.
I've had some time to do more measurements, the way you described in the post quoted above. I've measured OPA2134PA, NJM2068D and the good old NE5532AP.
This time, like last time around, I measured the input value as described. They were all the same, at 6.7 mV. That's 2.54 dBV to subtract from the total noise power.
I measured the total noise, A-weighted, but this time I used averaging. The numbers below are an average of five measurements. (Edit: Whenever the measurements fluctuated, I chose the worst result. The NJM2868D, for instance, showed down to 88.4 dBV noise.) So without further ceremony, here are my results:
(OP AMP: "Noise Power" - 2.54 dB = SNR)
OPA2134PA: 81.7 - 2.54 = 79.16 dB
NJM2068D: 88.0 - 2.54 = 85.46 dB
NE5532AP: 85.0 - 2.54 = 82.46 dB
This is done in daylight, which means no LED lights (or any other light sources) were turned on. All other equipment except for my computer and the QA400 were turned off.
Last edited:
Hi,
With the input short cicuited, current noise from the Op-Amp does not play a role, but with the cart attached, things will be rather different with a bipolar amp.
I'm glad that you achieved the same 79dB as my simulation gave. Remember that with cart attached this value drops to 77dB.
With the NE5534, short circuited gives 83dB but with the Cart attached this drops to 78dB.
No significant difference with the OPA 2134, which to my opinion is a better sounding Op-Amp.
The other reason to use a FET Op-Amp is to prevent DC current from flowing into the Cart.
The NE5534 can have an Ibias up to 1.5 uA, which is way too much for an MM cartridge.
If you want for whatever reason the best possible SNR, take the OPA1642 which is still affordable, a dual Op-Amp having a FET input with 5.5nV/rtHz.
But keep in mind that a SNR above 75dB is not contributing to a better phono-preamp.
Better to select the best sounding Op-Amp instead.
Hans
Just a follow-up question, why are bipolar op amps considered better for moving coil cartridges?
Lower input noise voltage. Since the MC DC resistance and inductance is very low (10's of ohms), the noise voltage of the bipolar plays only a small part. The noise current is then the dominant noise mechanism.
You can read about this stuff here:- http://hifisonix.com/riaa-equalizers/
There are some simulations which back up Hans's findings as well.
You can read about this stuff here:- http://hifisonix.com/riaa-equalizers/
There are some simulations which back up Hans's findings as well.
Last edited:
Bonsai,
You probably made a small typo, for an MC noise voltage plays a dominant role and noise current is insignificant because of the very low Cart resistance.
That's why you can use op-amp's with only 0.9nV/rtHz like the AD797, but with a noise current of 2pA/rtHz that would be detrimental for an MM.
Fet Op-Amps don't go much lower as 4nV/rtHz, so the choice is easy in favour of the Bipolar.
Hans
Since this is a phono stage this thread really ought to be in analog source with its cousins. Will move it there.Since a picture tells more than a thousand words, I took a LT1115 op-amp, having 0.9nV/rtHz and 1.2pA/rtHz, even a bit better than the AD797.
Go back to posting #80 and look at the two noise graphs, one short circuited and the other with Cart connected giving resp 76uV an 94uV noise.
Now look how the noise graphs are with the LT1125, again resp. short circuited and with Cart.
Noise in both cases is now resp 27uV and 234uV or the equivalent of resp 88dB and 69dB SNR A-weighted ref 5mV@1Khz.
This will make it quite clear why a Fet Op-Amp is by far the better choice for an MM.
Hans
Go back to posting #80 and look at the two noise graphs, one short circuited and the other with Cart connected giving resp 76uV an 94uV noise.
Now look how the noise graphs are with the LT1125, again resp. short circuited and with Cart.
Noise in both cases is now resp 27uV and 234uV or the equivalent of resp 88dB and 69dB SNR A-weighted ref 5mV@1Khz.
This will make it quite clear why a Fet Op-Amp is by far the better choice for an MM.
Hans
This last graph really illustrates your point Hans. That's quite a difference with the cartridge connected.
Here's the OPA2134 measurement graph btw. There's a peak at ~150 Hz that all but went away when I disconnected my laptop's power supply. It didn't affect the total noise power though, which (even with an average of 20 samples) fluctuated from -81.6 to -81.8 dBV and stayed at -81.7 most of the time.
Here's the OPA2134 measurement graph btw. There's a peak at ~150 Hz that all but went away when I disconnected my laptop's power supply. It didn't affect the total noise power though, which (even with an average of 20 samples) fluctuated from -81.6 to -81.8 dBV and stayed at -81.7 most of the time.
I see peaks at all uneven harmonics up to at least 650Hz.
This points in the direction of a 50Hz rectifier starting and stopping to conducting each time in the top of the 50Hz Hz sinus.
My guess is that the supply of your muffsy can be improved quite a bit.
To see how much better it can be, try to temporarily replace the power supply with two 9V batteries and measure again.
Hans
This points in the direction of a 50Hz rectifier starting and stopping to conducting each time in the top of the 50Hz Hz sinus.
My guess is that the supply of your muffsy can be improved quite a bit.
To see how much better it can be, try to temporarily replace the power supply with two 9V batteries and measure again.
Hans
It's the QA400 picking up noise from the laptop's power supply. The peaks are the same even if I disconnect the cable from the phono stage to the QA400.
Here's a screenshot I took with the NJM2068D in the phono stage, laptop PSU disconnected:
Here's a screenshot I took with the NJM2068D in the phono stage, laptop PSU disconnected:
Then find how you need to screen your test leads to clear away the interference.
Try a steel/iron box around the DUT and then a screen connected to the DUT box and to the QA box placed around the test leads.
Try a steel/iron box around the DUT and then a screen connected to the DUT box and to the QA box placed around the test leads.
Thanks AndrewT,
I suspect that the noise is injected in the QA400 through my laptop's USB interface. Although it's always worse at night when it's dark, even if I turn off all the LED lighting.
But you're quite right, I do need to be able to make repeatable measurements. I'm going on vacation, so it has to wait for a while.
I suspect that the noise is injected in the QA400 through my laptop's USB interface. Although it's always worse at night when it's dark, even if I turn off all the LED lighting.
But you're quite right, I do need to be able to make repeatable measurements. I'm going on vacation, so it has to wait for a while.
Here's the same measurement as yesterday, with the old Dell charger disconnected from mains power. The QA400 only shows one decimal, so it's not clear if it changed a full 0.3 dBV.
I left the aluminum enclosure open last night, and now I closed and screwed it shut.
I left the aluminum enclosure open last night, and now I closed and screwed it shut.
- Status
- Not open for further replies.