I now have a tubed phono pre-amp, but am not completely satisfied by the sound. Very warm sounding, but also sometimes blurry around tones, and distorted compared to other sources due to a lesser harmonic distorsion, eg tuner and CD.
This could be because the phono is a single ended design (2 ECC83 halves with passive Riaa inbetween) The line amp is an 6N1P SRPP, the poweramp E88CC SRPP and cap coupled to two good old & big Hitachi TO-3 powerfets per channel.
All Power supplies are single and with a choke, only the fets have a balanced PS, so in all stages the second harmonic is emphasised, except in the last Fet-stage, wherein the uneven harmonics are amplified due to the balanced design. There is no overall feedback whatsover, only local feedback on some places, a decoupling cap over cathode resistor.
So now i have 3 "even" stages in my ampl. chain and 1 "uneven"
I am thinking of a balanced fet phono MC\MM phono stage to replace the tubed phono. This to combine "uneven harmonic" amplifier stages with "even harmonic stages" to reduce overall distorsion. Even harmonics are better to listen to, but it still is distorsion.
What do you here thinks of this?
This could be because the phono is a single ended design (2 ECC83 halves with passive Riaa inbetween) The line amp is an 6N1P SRPP, the poweramp E88CC SRPP and cap coupled to two good old & big Hitachi TO-3 powerfets per channel.
All Power supplies are single and with a choke, only the fets have a balanced PS, so in all stages the second harmonic is emphasised, except in the last Fet-stage, wherein the uneven harmonics are amplified due to the balanced design. There is no overall feedback whatsover, only local feedback on some places, a decoupling cap over cathode resistor.
So now i have 3 "even" stages in my ampl. chain and 1 "uneven"
I am thinking of a balanced fet phono MC\MM phono stage to replace the tubed phono. This to combine "uneven harmonic" amplifier stages with "even harmonic stages" to reduce overall distorsion. Even harmonics are better to listen to, but it still is distorsion.
What do you here thinks of this?
Dear friend,
this is the classic monotube preamp.
Some times ago there was a beautiful article on Audiophile written by Hiraga whrw you should understand how much difficult is build a phono pre with a single tube.
The distorsion is probably due to need to have maximun gain from tubes.Much gain mean much distorsion and consider tou throw gain with passive eq.Simply add another tube e retouch the gain in the stages.
You should have a fast confirmation of this removing the passive riaa eq and use a feedback riaa.
Ciao
cicero32
this is the classic monotube preamp.
Some times ago there was a beautiful article on Audiophile written by Hiraga whrw you should understand how much difficult is build a phono pre with a single tube.
The distorsion is probably due to need to have maximun gain from tubes.Much gain mean much distorsion and consider tou throw gain with passive eq.Simply add another tube e retouch the gain in the stages.
You should have a fast confirmation of this removing the passive riaa eq and use a feedback riaa.
Ciao
cicero32
Dear tubee,
If you want use passive riaa it's better have a coumpound stage and not a single tube.The first stage must have high gain, high acceptance and low noise.All request are conflicting! In your design you should have 30- 35 dB in first stage and you throw it away with passive Riaa (-30dB).Try the feedback riaa, or feedback tout court. If pre sound better this is reason of your muddy sound and you can consider what configuration you want.
cascode or SRPP - passive RIAA - cascode or SRPP.
ciao cicero
P.S. I apologize for my poor english linguage
If you want use passive riaa it's better have a coumpound stage and not a single tube.The first stage must have high gain, high acceptance and low noise.All request are conflicting! In your design you should have 30- 35 dB in first stage and you throw it away with passive Riaa (-30dB).Try the feedback riaa, or feedback tout court. If pre sound better this is reason of your muddy sound and you can consider what configuration you want.
cascode or SRPP - passive RIAA - cascode or SRPP.
ciao cicero
P.S. I apologize for my poor english linguage
Hello all
Nice to have so much reactions!
The design-goals of the phono was to build it as simple as possible. They wanted a srpp first, but choosed a single stage. The first stage, 1st halve of ECC83 (Old Philips tubes) amplifies about 35dB, then a cap + resistor and one cap for RIAA (thats all!) and then 2nd halve which does about 25-30 dB to amplify the riaa loss. ( i fitted a decoupling cap over both cathode resistors to get some extra dB's, i have a Benz Gold with moderate output. The designer advised a cap for first stage only, a cap for 2nd stage only with low output carts.
I think i fitted Elna cerafines, not sure.
The line stage is SRPP, as the first stage of the hybrid poweramp. This is the reason why i don't want another srpp for the phono. A common cathode would be better i think, if this is possible for a phonostage.
Another fact: i have no feedback at all, not in phono, pre stage nor hybrid amp. I could try a small feedback in the phono stage and see what happens. Also a Sovtek 12AX7 could give some extra, but tried Haltron's no improvement, so the Sovtek won't either i think.
Nice to have so much reactions!
The design-goals of the phono was to build it as simple as possible. They wanted a srpp first, but choosed a single stage. The first stage, 1st halve of ECC83 (Old Philips tubes) amplifies about 35dB, then a cap + resistor and one cap for RIAA (thats all!) and then 2nd halve which does about 25-30 dB to amplify the riaa loss. ( i fitted a decoupling cap over both cathode resistors to get some extra dB's, i have a Benz Gold with moderate output. The designer advised a cap for first stage only, a cap for 2nd stage only with low output carts.
I think i fitted Elna cerafines, not sure.
The line stage is SRPP, as the first stage of the hybrid poweramp. This is the reason why i don't want another srpp for the phono. A common cathode would be better i think, if this is possible for a phonostage.
Another fact: i have no feedback at all, not in phono, pre stage nor hybrid amp. I could try a small feedback in the phono stage and see what happens. Also a Sovtek 12AX7 could give some extra, but tried Haltron's no improvement, so the Sovtek won't either i think.
I think the problem with your existing phono stage is not that it is unbalanced or that it is valve or that it uses passive equalisation (I use all three of these things), the problem is the ECC83. ECC83 has high ra (typically about 80k or so) so it has to be included in the design of the passive RIAA. Trouble is, it forms a rather large proportion of the series resistance in the RIAA network but it varies. It is most likely that you have inaccurate RIAA and that is why it sounds lush.
It may also be that the RIAA stage is distorting due to inadequate HT voltage or poor gain structure, but without a diagram it's not possible to comment fully.
It may also be that the RIAA stage is distorting due to inadequate HT voltage or poor gain structure, but without a diagram it's not possible to comment fully.
A high gain circuit is normally required, as the input level is very low and the passive networks reduces effectively most frequencies.
1) As the RIAA filter input impedance loads the first stage, the RIAA filter input impedance should be as HIGHER as possible with respect to tube input impedance.
2) Anyway, as any resistance introduces a noise and higher the resistance value, higher the noise level, it is mandatory to keep all resistors values as LOW as possible.
If you design a RIAA stage, you'll find out that it is not at all easy to select commercially available component values. Often you have to use more than one component in parallel or series to build up the required value
As the design is to reproduce very low frequencies, and if high accuracy is required in this area, then it is necessary to take into account any bypass or coupling capacitors in the system...
Noise reduction is a major issue
The first three points make it necessary to choose a high gain, low internal impedance tube for the first stage... and if you look through data sheets for such a tube you'll really be surprised to find such a small number of tubes with these features.
The search leads to a couple of well-known tubes, which have relatively low internal impedance and relatively high gain: their selection too is therefore the result of a compromise.
The two tubes are
5842/417A. A very discussed tube without high linearity but other good qualities that make it suitable to use in the first stage (very low level) of the phono module. Some versions can be subject to severe microphony problems, others can be noisy; as far as I know it is no longer in production and can be found only as NOS. It also has a rather short life and EACH tube have
different parameters, a selection is due.
ECC88/6922 tube family. Offers very good linearity; here too some versions can be subject to bad microphony problems, others can be noisy; it is still in production, even in high quality, low noise versions, and there is a relatively large amount of old, very high quality tubes available; life is rather long and costs are not enormous.
The first stage should been biased in order to get a low anode voltage and a rather high current through them. This working point has a lower internal impedance. This reduces the dependency of the RIAA filter's frequency response on the tubes characteristics. Also reduced is tube thermal noise induced by internal resistance. Even more important for tube noise reduction is the low supply voltage.
In your pre design, a 0,4 mV (1kHZ)input signal is amplified up to 23 mV at the output of first stage.
The passive RIAA network cuts it down by about 30dB, so that we find 0,7 mV at the input of second stage, which very roughly shows that the second stage is only 8dB less sensitive then first stage, to tube noise around 1 kHz. In fact, I had to modify the design of second stage after testing for noise problems, reducing the anodic voltage.<br> Anyhow, in order to have the same pre output level as above at 20Hz, an input roughly of the order of 0.4mV is required. On the other hand, in order to have the same pre output level at 20kHz, an input very roughly of the order of 40mV is required. If we suppose an even first stage noise distribution through the full audio frequency range the signal-to-noise ratio at low frequencies is far worse than high frequencies. From a practical point of view, the phono input and first stage thermal noise appears as an essentially low frequency hiss and rumble.
I suppose the voltage of your first stage be around 200 - 250 V,
the Rp is 62,5 kOhm (80 kOhm @100V) so no low impedence RIAA
= noisy stage.
Your input from Benz is too low (it's a MC) = probable signal compression and/or overload on high frequency side.
Those are, in my opinion the reasons of the poor sound of your pre.
ciao cicero32
1) As the RIAA filter input impedance loads the first stage, the RIAA filter input impedance should be as HIGHER as possible with respect to tube input impedance.
2) Anyway, as any resistance introduces a noise and higher the resistance value, higher the noise level, it is mandatory to keep all resistors values as LOW as possible.
If you design a RIAA stage, you'll find out that it is not at all easy to select commercially available component values. Often you have to use more than one component in parallel or series to build up the required value
As the design is to reproduce very low frequencies, and if high accuracy is required in this area, then it is necessary to take into account any bypass or coupling capacitors in the system...
Noise reduction is a major issue
The first three points make it necessary to choose a high gain, low internal impedance tube for the first stage... and if you look through data sheets for such a tube you'll really be surprised to find such a small number of tubes with these features.
The search leads to a couple of well-known tubes, which have relatively low internal impedance and relatively high gain: their selection too is therefore the result of a compromise.
The two tubes are
5842/417A. A very discussed tube without high linearity but other good qualities that make it suitable to use in the first stage (very low level) of the phono module. Some versions can be subject to severe microphony problems, others can be noisy; as far as I know it is no longer in production and can be found only as NOS. It also has a rather short life and EACH tube have
different parameters, a selection is due.
ECC88/6922 tube family. Offers very good linearity; here too some versions can be subject to bad microphony problems, others can be noisy; it is still in production, even in high quality, low noise versions, and there is a relatively large amount of old, very high quality tubes available; life is rather long and costs are not enormous.
The first stage should been biased in order to get a low anode voltage and a rather high current through them. This working point has a lower internal impedance. This reduces the dependency of the RIAA filter's frequency response on the tubes characteristics. Also reduced is tube thermal noise induced by internal resistance. Even more important for tube noise reduction is the low supply voltage.
In your pre design, a 0,4 mV (1kHZ)input signal is amplified up to 23 mV at the output of first stage.
The passive RIAA network cuts it down by about 30dB, so that we find 0,7 mV at the input of second stage, which very roughly shows that the second stage is only 8dB less sensitive then first stage, to tube noise around 1 kHz. In fact, I had to modify the design of second stage after testing for noise problems, reducing the anodic voltage.<br> Anyhow, in order to have the same pre output level as above at 20Hz, an input roughly of the order of 0.4mV is required. On the other hand, in order to have the same pre output level at 20kHz, an input very roughly of the order of 40mV is required. If we suppose an even first stage noise distribution through the full audio frequency range the signal-to-noise ratio at low frequencies is far worse than high frequencies. From a practical point of view, the phono input and first stage thermal noise appears as an essentially low frequency hiss and rumble.
I suppose the voltage of your first stage be around 200 - 250 V,
the Rp is 62,5 kOhm (80 kOhm @100V) so no low impedence RIAA
= noisy stage.
Your input from Benz is too low (it's a MC) = probable signal compression and/or overload on high frequency side.
Those are, in my opinion the reasons of the poor sound of your pre.
ciao cicero32
Thanks EC 8010 and Cicero32 for very elaborate explaining of possibe problems!
I have been digging in 1964 Philips datasheets of ECC83, this tube will give lesser distortion figures on higher supply voltages (300V) The first stage runs at about 210V, the second only on about 120V. So they might work now in a more non lineair part of I/V diagram.
No scanner online, but schematic is simple:
MD output ->100k to gnd (so to big for the Benz) , 100pF to Gnd(40000Hz filter),0.68uF coupling cap(rumble filter),100k to Gnd, 1k miller resistor to grid of 1st 1/2 ECC83.
VCC 210V, R anode is 100K, R cathode 900 ohm/220uF Panasonic/0.1uF Mkp.
From anode 1st stage 0.47uF coupling cap, 68.1k series resistor, after that a 22N low freq. RIAA with 14k in series to Gnd, after that 7800pF high freq. roll-off RIAA to Gnd. This is followed by a 1M to gnd (leaking R for cap?), and a 1k miller R in to grid of 2nd stage. The 2nd 1/2 ECC83 runs on about 120V, with 20k anode R, 600 ohm cathode R/220uF Panasonic/0.1uF MKP.
The anode of 2nd stage is coupled to line stage(pot/switch) with 0.47uF coupling cap/475k to Gnd.
I have mixed the types of coupling caps: rumble is Wima mkp, the two coupling caps are Wima MKS, and Philips MKC. This improved the sound but only a little. (First all coupling caps where MKP (Ero + Wima))
The RIAA is in polystyrene (added different caps to get the right value)
The Phono stage design is from a 1994 magazine, and i built it in free air PTP few years ago, they recommended/sold a pcb for it.
The Benz gives 0.4mV at 3.54 cm/s rms, rec. loading 200 to 47000 ohm.
So problably the phono stage will give less distorsion with higher supply voltage, but noise will increase too then. Maybe some mods of values?
I saw somewhere else here on diy a RIAA calculator.
Btw, my wife went to bed after she saw this typing, she knows me by now...
I have been digging in 1964 Philips datasheets of ECC83, this tube will give lesser distortion figures on higher supply voltages (300V) The first stage runs at about 210V, the second only on about 120V. So they might work now in a more non lineair part of I/V diagram.
No scanner online, but schematic is simple:
MD output ->100k to gnd (so to big for the Benz) , 100pF to Gnd(40000Hz filter),0.68uF coupling cap(rumble filter),100k to Gnd, 1k miller resistor to grid of 1st 1/2 ECC83.
VCC 210V, R anode is 100K, R cathode 900 ohm/220uF Panasonic/0.1uF Mkp.
From anode 1st stage 0.47uF coupling cap, 68.1k series resistor, after that a 22N low freq. RIAA with 14k in series to Gnd, after that 7800pF high freq. roll-off RIAA to Gnd. This is followed by a 1M to gnd (leaking R for cap?), and a 1k miller R in to grid of 2nd stage. The 2nd 1/2 ECC83 runs on about 120V, with 20k anode R, 600 ohm cathode R/220uF Panasonic/0.1uF MKP.
The anode of 2nd stage is coupled to line stage(pot/switch) with 0.47uF coupling cap/475k to Gnd.
I have mixed the types of coupling caps: rumble is Wima mkp, the two coupling caps are Wima MKS, and Philips MKC. This improved the sound but only a little. (First all coupling caps where MKP (Ero + Wima))
The RIAA is in polystyrene (added different caps to get the right value)
The Phono stage design is from a 1994 magazine, and i built it in free air PTP few years ago, they recommended/sold a pcb for it.
The Benz gives 0.4mV at 3.54 cm/s rms, rec. loading 200 to 47000 ohm.
So problably the phono stage will give less distorsion with higher supply voltage, but noise will increase too then. Maybe some mods of values?
I saw somewhere else here on diy a RIAA calculator.
Btw, my wife went to bed after she saw this typing, she knows me by now...
I found some phono stages with ECC83, in the book: Tube amplifiers for audio and guitars. They advised a combination of passive/active RIAA correction. The low frequency part is active.
EC8010, you have a similar design, passive with ECC83. Do you have a schematic of it to compare with mine?
Thanks.
EC8010, you have a similar design, passive with ECC83. Do you have a schematic of it to compare with mine?
Thanks.
Stranges operating points, seems to be driver stadium of classic active feedback preamps.In other word this pream seem be two driver stages with a passive riaa in between.The stages aws high gain high distorsion but with feedback this was not a problem.
There are many moder "tubes guru design" selling projects or hardware (with great reputation in HiFi world) who have NOT IDEA of what is a tube design.You can this kind of things in first Michaelson & Austin products or Quicksilver where you have half tube doing nothing.
An example is the Williamson circuit, a kind of magic this is the first HiFi poweramp, despite the poor laminated output transformer and winding not "esoteric" , the normal dispersion parameters of components and tubes; this ampli was stable on all loads and was reproducible in all parts of the worlds.This was due to compensating poles and zeroes with RC groups.
Naturally the Modern Hifi Guru don't know Bode & Nyquist work and rework the Williamson circuit in no feedback ultrawide amplifier! I've seen Williamson transformed in Monotriode with an indirect heating tube! It's not a joke ! You can see it on Costruire Hi Fi a other times famous italian tube's Hifi Journal (with most
better engenners).
This thing can't sound good!
You could assemble an AntiRiaa circuit and put it in front of pre,
with a generator and a scope you see what happen.
If you have't generator e/o scope you can listen a linear source like FM radio or AUX and listen.This is real sound of your Pre.
ciao cicero32
There are many moder "tubes guru design" selling projects or hardware (with great reputation in HiFi world) who have NOT IDEA of what is a tube design.You can this kind of things in first Michaelson & Austin products or Quicksilver where you have half tube doing nothing.
An example is the Williamson circuit, a kind of magic this is the first HiFi poweramp, despite the poor laminated output transformer and winding not "esoteric" , the normal dispersion parameters of components and tubes; this ampli was stable on all loads and was reproducible in all parts of the worlds.This was due to compensating poles and zeroes with RC groups.
Naturally the Modern Hifi Guru don't know Bode & Nyquist work and rework the Williamson circuit in no feedback ultrawide amplifier! I've seen Williamson transformed in Monotriode with an indirect heating tube! It's not a joke ! You can see it on Costruire Hi Fi a other times famous italian tube's Hifi Journal (with most
better engenners).
This thing can't sound good!
You could assemble an AntiRiaa circuit and put it in front of pre,
with a generator and a scope you see what happen.
If you have't generator e/o scope you can listen a linear source like FM radio or AUX and listen.This is real sound of your Pre.
ciao cicero32
tubee said:
Hi JW,
Maybe a better idea is to download ARTA which is an excellent set of audio analysis tools for use with a soundcard. If you have a 96 kHz soundcard you can measure up to almost 48 kHz. The free evaluation version is fully functional except for saving.
If you use the RIAA curve for the mic compensation you can measure pretty good the RIAA of your phono pre without an inverted RIAA hardware filter and probably much more accurate.
If you are interested I have mic compensation files available with the RIAA curve, both nominal and with te 3.18 us Neumann time constant.
Cheers
In my opinion an antiRiaa with high precision components is better because with the mic type you have to flat compensate mike in advance and after you can equilize with nominal value;
otherwhise you have great summing errors at extreme low - high part of the curve.You can't use a chip mike, the hardware anti RIAA need only some 1% or better resistor and some high quality cap.You should select capacitors and you have a rig for the homelab.Its a very piece of high precision and you can use it for fast on field evaluation or for more accurate measurements.
I think you have high distorsion values and badly accurate riaa,
a serious riaa from low point to high point have a 40 dB of difference.If your first stage have a 35 dB@1kHz gain on some frequency you should have -5dB gain!
Many peoples use a passive riaa for a supposed audio superiority, why the most famous preamplifier are feedback equalised?
The mixed solution (passive and active) is a good solution with a good soundbecause it's an easy load to drive, you should split riaa net (active or passive) for same reason.
ciao cicero32
otherwhise you have great summing errors at extreme low - high part of the curve.You can't use a chip mike, the hardware anti RIAA need only some 1% or better resistor and some high quality cap.You should select capacitors and you have a rig for the homelab.Its a very piece of high precision and you can use it for fast on field evaluation or for more accurate measurements.
I think you have high distorsion values and badly accurate riaa,
a serious riaa from low point to high point have a 40 dB of difference.If your first stage have a 35 dB@1kHz gain on some frequency you should have -5dB gain!
Many peoples use a passive riaa for a supposed audio superiority, why the most famous preamplifier are feedback equalised?
The mixed solution (passive and active) is a good solution with a good soundbecause it's an easy load to drive, you should split riaa net (active or passive) for same reason.
ciao cicero32
Hi cicero32,
The mic compensation file is just a file that corrects the measured response. So if you use the RIAA curve for it, well it just compensates for the RIAA correction of the pre amp giving a ruler flat final response (if the RIAA in your pre is accurate). Such a compensation file can be made to any accuracy with MS Excel using the RIAA time constants in a Laplace function.
You are right not to overdrive the pre at the low end and you should take care of the 20 dB boost at 20 Hz. So for a MC pre you should attenuate the soundcard’s output to say 0,1 mV or maybe lower by a resistor divider. Resolution is not an issue, a 16 bits soundcard has a dynamic range of over 90 dB that easily accommodates the 40 dB of the RIAA compensation. And if you use a 24 bit card like my M-Audio el cheapo Audiophile 2496 it is no problem at all.
Also noise is no problem. Since it is random in nature you get rid of it by averaging. In ARTA you can also measure with a stepped sine excitation and the noise can be heavily suppressed by using a long enough integration time for the detector.
Cheers
The mic compensation file is just a file that corrects the measured response. So if you use the RIAA curve for it, well it just compensates for the RIAA correction of the pre amp giving a ruler flat final response (if the RIAA in your pre is accurate). Such a compensation file can be made to any accuracy with MS Excel using the RIAA time constants in a Laplace function.
You are right not to overdrive the pre at the low end and you should take care of the 20 dB boost at 20 Hz. So for a MC pre you should attenuate the soundcard’s output to say 0,1 mV or maybe lower by a resistor divider. Resolution is not an issue, a 16 bits soundcard has a dynamic range of over 90 dB that easily accommodates the 40 dB of the RIAA compensation. And if you use a 24 bit card like my M-Audio el cheapo Audiophile 2496 it is no problem at all.
Also noise is no problem. Since it is random in nature you get rid of it by averaging. In ARTA you can also measure with a stepped sine excitation and the noise can be heavily suppressed by using a long enough integration time for the detector.
Cheers
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.