Hello everyone. I ask you what opinion you have about phono preamps. In which of them the sound is more pleasant.
I have built several passive and active preamplifiers of the very named in this forum and in others (all solid state). I have compared with the preamplifier that owns the integral amplifier Marantz PM6006 (I do not know if it is active or passive) and subjectively the DIY have better detail.
My impression is that those who are passive sound more agreeable and those who are active have more dynamics. Is it like this or is it only a psychoacoustic or subjective effect? I accept opinions.
I have built several passive and active preamplifiers of the very named in this forum and in others (all solid state). I have compared with the preamplifier that owns the integral amplifier Marantz PM6006 (I do not know if it is active or passive) and subjectively the DIY have better detail.
My impression is that those who are passive sound more agreeable and those who are active have more dynamics. Is it like this or is it only a psychoacoustic or subjective effect? I accept opinions.
Based on some circuits I have built ( SS ) I prefer the passive filter circuits. I think the capacitors used in the RIAA filters also matter. Polypropylene over the usual polyester. However I haven't made any AB comparison to confirm this. Many polyester filters sound good too.
There should be a good overload margin too. Tubes excel in this generally. Signal to noise isn't that crucial I think because most decent circuits have significantly lower audible noise than the disc itself.
Some older discrete RIAA circuits had good ratings too. From NAD for example. They were all active feedback circuits. Maybe the active circuits sound different from the passive circuits due to the different distortion spectrum ( assuming the RIAA curve is accurate ). With varying load and gain in the feed back the distortion products for active RIAA should all over the place ? Not necessarily very high but audible ?
There should be a good overload margin too. Tubes excel in this generally. Signal to noise isn't that crucial I think because most decent circuits have significantly lower audible noise than the disc itself.
Some older discrete RIAA circuits had good ratings too. From NAD for example. They were all active feedback circuits. Maybe the active circuits sound different from the passive circuits due to the different distortion spectrum ( assuming the RIAA curve is accurate ). With varying load and gain in the feed back the distortion products for active RIAA should all over the place ? Not necessarily very high but audible ?
Hello everyone. I discovered new details and incredible dynamics with the acquisition of the integrated amplifier Marantz PM6006 (previously had Rotel RA970BX).
But the DIY preamplifier I own I like more than the one with this amp.
All have a spectacular sound, but different from each other.
They are constructed from kits. They are RJS VSPS 300, dual mono, active RIAA filter, 1% DALE resistors, Cornell Dubilier capacitors; Scrodahl Mufssy, passive RIAA filter based on AK hi end phono stage, 1% metal film resistors, Wima capacitors; Phono preamplifier by Rod Elliot, RIAA mixed filter; Solidphono of TNT, with passive RIAA filter, with first stage amplifier SSM2017, but with diy PCB, Wima capacitors and resistors 1% metallic film.
All with very good sound, but I repeat different between them.
I want you with more experience than I, to tell me which one fits the real sound better.
Greetings and I hope your opinions.
But the DIY preamplifier I own I like more than the one with this amp.
All have a spectacular sound, but different from each other.
They are constructed from kits. They are RJS VSPS 300, dual mono, active RIAA filter, 1% DALE resistors, Cornell Dubilier capacitors; Scrodahl Mufssy, passive RIAA filter based on AK hi end phono stage, 1% metal film resistors, Wima capacitors; Phono preamplifier by Rod Elliot, RIAA mixed filter; Solidphono of TNT, with passive RIAA filter, with first stage amplifier SSM2017, but with diy PCB, Wima capacitors and resistors 1% metallic film.
All with very good sound, but I repeat different between them.
I want you with more experience than I, to tell me which one fits the real sound better.
Greetings and I hope your opinions.
Hi,
I'm pretty sure that nobody could answer that question seriously.
There are too many factors involved that affect the sonic footprint.
Design of the power supplies, the topology and precision of the RIAA equalizing network, interchannel balance, design of the active stages, required gain, overload capability, the use and number of (global) NFB loops and how many stages the loops encompass, and not least parts quality.
A fully passive RIAA requires 20dB more gain than a split RIAA or a fully active RIAA.
That alone may lead to sonic differences, maybe due to different noise floor level and frequency response.
The probably most decisive factor is the precision of the equalization and the interchannel balance.
If the parts values aren't highly precise the deviations from a linear amplitude response can easily be several dB high.
Its not at all easy to guarantee less than 0.5dB deviation, from which sonic differences clearly become recognizable.
And let's not forget, that for a fair comparison the two contestands are required to be levelled in output volume.
Personally I like split RIAAs as they require less gain and its easier to design the networks highly precise.
It also allows to optimize parts values to improve on noise performance and they follow the RIAA precisely even above the audio band.
The only drawback is the requirement of at least a second active stage.
And lets under no circumstances forget the most important parameter ... Your personal taste.
The idea of what sounds 'real' varys more between individuals than a circuit designer could possibly achieve with changing part A against part B.
jauu
Calvin
I'm pretty sure that nobody could answer that question seriously.
There are too many factors involved that affect the sonic footprint.
Design of the power supplies, the topology and precision of the RIAA equalizing network, interchannel balance, design of the active stages, required gain, overload capability, the use and number of (global) NFB loops and how many stages the loops encompass, and not least parts quality.
A fully passive RIAA requires 20dB more gain than a split RIAA or a fully active RIAA.
That alone may lead to sonic differences, maybe due to different noise floor level and frequency response.
The probably most decisive factor is the precision of the equalization and the interchannel balance.
If the parts values aren't highly precise the deviations from a linear amplitude response can easily be several dB high.
Its not at all easy to guarantee less than 0.5dB deviation, from which sonic differences clearly become recognizable.
And let's not forget, that for a fair comparison the two contestands are required to be levelled in output volume.
Personally I like split RIAAs as they require less gain and its easier to design the networks highly precise.
It also allows to optimize parts values to improve on noise performance and they follow the RIAA precisely even above the audio band.
The only drawback is the requirement of at least a second active stage.
And lets under no circumstances forget the most important parameter ... Your personal taste.
The idea of what sounds 'real' varys more between individuals than a circuit designer could possibly achieve with changing part A against part B.
jauu
Calvin
Last edited:
Hi, A great question and as others have mentioned no clear answer. The one closest is that the best one is the one you like best. In my own experience in both diy and using commercial products it seems I prefer tube based passive preamps best. The next best is solid state IC based ones with active equalization. An odd pair for sure. The SS ones are typically quieter (one is at -106db) and while hard to measure more precise. The tube ones are more engaging to me. Go figure
Though I don't do it this way, the TI app note is interesting- http://www.ti.com/lit/an/snaa046a/snaa046a.pdf
I prefer active eq and precede the eq stage with a 3X buffer. That gives me some advantage in lowering resistors and keeping noise low. Self also gives some reasons why active is the preferred way to go in one of his books.
I prefer active eq and precede the eq stage with a 3X buffer. That gives me some advantage in lowering resistors and keeping noise low. Self also gives some reasons why active is the preferred way to go in one of his books.
Passive or active?
1, 2 or 3 gain blocks?
In what order?
Noise vs head-room?
Lately I have been pondering about these and other questions, and to answer at least some them I have made a LTSpice analysis (see attached schematic and ltspice file)
A few comments (in no particular order):
1) I have decided to use a total gain of 60dB (1000x) @ DC. A little higher than is the norm, but it brings the output closer to the output from say a CD-player.
2) I have used OP37 throughout, not out of preference, but simply because it comes with LTSpice. (don't use in active topologies, as it may become unstable)
3) Looking at circuit A) and B), then the B) version must be preferred, as more of the signal is amplified earlier= less noise 4) Circuit C) (which is found in the LM4562 datasheet) is really the worst of the lot, from app. 4.5kHz the input to the second amplifier is *lower* than the signal from the cartridge(!) i.e. there is a noise penalty!
5) My reasoning for using three buffers/gain blocks comes from the fact that if looking at OP amps specs, you will notice that the more gain is required, the more distortion you'll get. Keeping gain low (20dB) you can use modest (audio) OP amps e.g. LM833. I've even tried xx4558 and it actually sounded OK. I allways use xx4558 for testing as if they get zapped the loss is minimal.
6) About circuit D): I like this because of it's simplicity and minimal parts-count (for a three OP amp RIAA that is). The last amplifer takes care of the 3180 and 318muSec. The second takes care of the 75muSec, a sideeffect is that it also produces a 7.5muSec zero, hence the extra cap on the first, which then produces a pole at 7.5muSec (cancelling the former pole) and a zero at 0.75muSec (212kHz) which is of no consequence in my book.
7) I have trimmed the RIAA encoder, such that the final output is 1V - or 0dB in LTSpice.
8) About noise: as far as I can see, the only elements that matters are
A) The cartridge (by far)
B) the first resistor in in the various feedback networks (R15, R2, R24, R30
and R14)
9) The impedance level is high enough to allow just about any audio OPA -
from MC33078 to AD797!
I have not attached any curves, as there just is too many combinations.
Go play with them!
1, 2 or 3 gain blocks?
In what order?
Noise vs head-room?
Lately I have been pondering about these and other questions, and to answer at least some them I have made a LTSpice analysis (see attached schematic and ltspice file)
A few comments (in no particular order):
1) I have decided to use a total gain of 60dB (1000x) @ DC. A little higher than is the norm, but it brings the output closer to the output from say a CD-player.
2) I have used OP37 throughout, not out of preference, but simply because it comes with LTSpice. (don't use in active topologies, as it may become unstable)
3) Looking at circuit A) and B), then the B) version must be preferred, as more of the signal is amplified earlier= less noise 4) Circuit C) (which is found in the LM4562 datasheet) is really the worst of the lot, from app. 4.5kHz the input to the second amplifier is *lower* than the signal from the cartridge(!) i.e. there is a noise penalty!
5) My reasoning for using three buffers/gain blocks comes from the fact that if looking at OP amps specs, you will notice that the more gain is required, the more distortion you'll get. Keeping gain low (20dB) you can use modest (audio) OP amps e.g. LM833. I've even tried xx4558 and it actually sounded OK. I allways use xx4558 for testing as if they get zapped the loss is minimal.
6) About circuit D): I like this because of it's simplicity and minimal parts-count (for a three OP amp RIAA that is). The last amplifer takes care of the 3180 and 318muSec. The second takes care of the 75muSec, a sideeffect is that it also produces a 7.5muSec zero, hence the extra cap on the first, which then produces a pole at 7.5muSec (cancelling the former pole) and a zero at 0.75muSec (212kHz) which is of no consequence in my book.
7) I have trimmed the RIAA encoder, such that the final output is 1V - or 0dB in LTSpice.
8) About noise: as far as I can see, the only elements that matters are
A) The cartridge (by far)
B) the first resistor in in the various feedback networks (R15, R2, R24, R30
and R14)
9) The impedance level is high enough to allow just about any audio OPA -
from MC33078 to AD797!
I have not attached any curves, as there just is too many combinations.
Go play with them!
Attachments
TI is somewhat like the Soviet Union of yore -- authors are dropped from the application note and removed from institutional memory after they've been laid off.
Just sayin'
I have used an Allen Wright FVP pre-amp for many years (purchased in 1987). Active RIAA eq, with amplification provided by valves. I enjoy the sound of it very much. But I haven't really used it since I moved to strain gauge cartridges in 2012.
I recently bought a LO moving coil. Since I need more gain, and my system is all balanced, I built a new RIAA amp. Passive RIAA, all active devices are BJT, power supply is SLA battery. I enjoy the sound of it very much.
So which is better? Active or passive? There are so many variables that it is impossible to answer this question. Either way can provide excellent results. However, on balance, I prefer my WIN strain gauge cartridge. Because the best RIAA is no RIAA!!!
I recently bought a LO moving coil. Since I need more gain, and my system is all balanced, I built a new RIAA amp. Passive RIAA, all active devices are BJT, power supply is SLA battery. I enjoy the sound of it very much.
So which is better? Active or passive? There are so many variables that it is impossible to answer this question. Either way can provide excellent results. However, on balance, I prefer my WIN strain gauge cartridge. Because the best RIAA is no RIAA!!!
Not all the records are the same .Some put out 30% more S/N ratio than others.Some cartridges get the best possible out of anything.The first stage of a phono preamp would better be able to deal with highest possible gain and both low and high S/N ratio of the vinil.I built a complex passive riaa in the way that one of the two switchable passive network i used looked pretty much the same as the worse Texas Instruments C version, but i had a gain of approximately 40 in the first stage which was a hybrid- tube /silicon transistor stage , while the second stage raised the signal up to maximum 4 volts. used an eliptical stylus on mm old sansui cartridge as refference and that cartridge favoured the amplification of high ferquencies.Still the sound was very dynamic.Also i used two capacitors to couple between stages and output.
BUT, the measurements were done extensively from 10 hz up to 480 khz sinus and sqw linearity was best between 500hz and 25khz /1.1 kohm output load with one atiriaa network and 100hz...10khz with another .
This was my first stage and measurements
unfortunately it's in romanian...)
UNU-preamplificatorul pentru pick-up - Page 3 - Audio pe tuburi - Forumul Electronistilor
my antiriaa filter for calibration
http://i60.tinypic.com/2j35rep.jpg
my both passive riaa networks:
http://i57.tinypic.com/288m7tc.jpg
Both riaa networks measured the same but sound differently.The lowest impedance one sounded more flat , i would say linear, but the first was more pleasant and sounded more ""dynamic", whatever that might have mean...Allthough the sqw for under 500hz is not great it didn't seem to me like having less or weak bass response.SQW and capacitive coupling is not the best way of measuring bass response.I did sinus measurements but they looked all the same, i mean good enough from 10hz to 480 khz.
I listened once to a good DENON poa preamp in a fairly good setup, but
I didn't try many values for capacitive compensation of the cartridge, just the classical 100pf(silver mica).
I could not complain of the quality of the capacitors...i used the best possible all throughout riaa network and coupling.
i think that i can make some real improovements to this circuit but the sound was really good and i compared that sound to a lot of other active preamps.Unfortunately i didn't came across the best of active riaa and if i am really curious how the active phono preamp found in Kenwood L-02A sounds , which i think should be the best in active riaa world.I wanted to build it as a stand alone unit myself, but i didn't have the time and resources for a long time.
I would really like to try all the versions described by @LLPe but in balanced configuration with ne5534 as i had a handfull of original signetics.
BUT, the measurements were done extensively from 10 hz up to 480 khz sinus and sqw linearity was best between 500hz and 25khz /1.1 kohm output load with one atiriaa network and 100hz...10khz with another .
This was my first stage and measurements
unfortunately it's in romanian...)UNU-preamplificatorul pentru pick-up - Page 3 - Audio pe tuburi - Forumul Electronistilor
my antiriaa filter for calibration
http://i60.tinypic.com/2j35rep.jpg
my both passive riaa networks:
http://i57.tinypic.com/288m7tc.jpg
Both riaa networks measured the same but sound differently.The lowest impedance one sounded more flat , i would say linear, but the first was more pleasant and sounded more ""dynamic", whatever that might have mean...Allthough the sqw for under 500hz is not great it didn't seem to me like having less or weak bass response.SQW and capacitive coupling is not the best way of measuring bass response.I did sinus measurements but they looked all the same, i mean good enough from 10hz to 480 khz.
I listened once to a good DENON poa preamp in a fairly good setup, but
I didn't try many values for capacitive compensation of the cartridge, just the classical 100pf(silver mica).
I could not complain of the quality of the capacitors...i used the best possible all throughout riaa network and coupling.
i think that i can make some real improovements to this circuit but the sound was really good and i compared that sound to a lot of other active preamps.Unfortunately i didn't came across the best of active riaa and if i am really curious how the active phono preamp found in Kenwood L-02A sounds , which i think should be the best in active riaa world.I wanted to build it as a stand alone unit myself, but i didn't have the time and resources for a long time.
I would really like to try all the versions described by @LLPe but in balanced configuration with ne5534 as i had a handfull of original signetics.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.