The cartridge i use is the Ortofon blue,
Specifications
•Output voltage at 1000 Hz, 5cm/sec: 5.5 mV
•Channel balance at 1 kHz: 1.5 dB
•Channel separation at 1 kHz: 25 dB
•Channel separation at 15 kHz: 15 dB
•Frequency range at - 3dB: 20-25.000 Hz
•Frequency response: 20-20.000 Hz + 2 / - 1 dB
•Tracking ability at 315Hz at recommended tracking force: 80 µm
•Compliance, dynamic, lateral: 20 µm/mN
•Stylus type: Nude Elliptical
•Stylus tip radius: r/R 8/18 µm
•Tracking force range: 1,6-2,0g (16-20 mN)
•Tracking force, recommended: 1,8 g (18 mN)
•Tracking angle: 20°
•Internal impedance, DC resistance: 1,3 kOhm
•Internal inductance: 700 mH
•Recommended load resistance: 47 kOhm
•Recommended load capacitance: 150-300 pF
•Cartridge weight: 7.2 g
The next audio part is my EF6 triode mode parafeed preamp with a input impedance of 47 kohm
Specifications
•Output voltage at 1000 Hz, 5cm/sec: 5.5 mV
•Channel balance at 1 kHz: 1.5 dB
•Channel separation at 1 kHz: 25 dB
•Channel separation at 15 kHz: 15 dB
•Frequency range at - 3dB: 20-25.000 Hz
•Frequency response: 20-20.000 Hz + 2 / - 1 dB
•Tracking ability at 315Hz at recommended tracking force: 80 µm
•Compliance, dynamic, lateral: 20 µm/mN
•Stylus type: Nude Elliptical
•Stylus tip radius: r/R 8/18 µm
•Tracking force range: 1,6-2,0g (16-20 mN)
•Tracking force, recommended: 1,8 g (18 mN)
•Tracking angle: 20°
•Internal impedance, DC resistance: 1,3 kOhm
•Internal inductance: 700 mH
•Recommended load resistance: 47 kOhm
•Recommended load capacitance: 150-300 pF
•Cartridge weight: 7.2 g
The next audio part is my EF6 triode mode parafeed preamp with a input impedance of 47 kohm
OK, that narrows things down. You'll want an MM stage with low input capacitance. That means a cascode or (at the least) a first stage with a lower mu triode than 6SL7 or 12AX7. And you'll want something for which there are circuit boards. Allen Wright's preamps are a good choice- the FVP would do well for you. Ditto John Broskie's Aikido phono. There's probably more.
Your cathode follower will work fine into a 47k load- your problem is RIAA conformance and input capacitance.
Your cathode follower will work fine into a 47k load- your problem is RIAA conformance and input capacitance.
I have built a phono amp and it doesm not sound as I suspected and as promisedthe high frequencies sound compressed and there is no air around the instruments. Are there comments or remarks about the scematic that I use?
I'm not surprised: that design is hideous. The first 6SL7 is loaded way too heavily. The design nominal plate resistance for this type is 44K, and with a 100K plate resistor, and the RIAA network, the effective plate load is right around 62K. Triodes like very light loading in order to operate in a linear manner. That would be bad enough, but the cathode degeneration drives the effective plate resistance even higher, making the loading that much worse.
The second 6SL7 stage still has a barely adequate 100K plate load, and that would be the case without cathode degeneration. Adding in the extra plate resistance makes that 100K far too small for this type. I've used 6SL7s before, and I like a much lighter plate load. That's simply the nature of the beast: the lighter the plate load, the better the linearity to the practical maximum of CCS plate loading. The 6SL7 is one of the better audio triodes, but it doesn't work magic, and it can not compensate for a fundamentally bad design.
This was designed by someone who really didn't know what he was doing. There is a lot of that going around these days, and tubes aren't transistors. Unlike transistors (or vacuum tube pents) what you hang on the cathode affects the plate characteristics (and vise versa). Here, the designer overlooked that detail.
I'd give that design
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.
An externally hosted image should be here but it was not working when we last tested it.
out of five.
One gadget that you'll find essential for making a phono stage is an inverse RIAA network. This allows you to test your results within the limitations of ordinary test equipment, even a computer soundcard, if you're lucky. There are a good variety of likely candidates in passive circuits fed at roughly line level in, phono out. If you make two (for stereo) you can do a listening test, with the inverse network and phono equalizer vs. without, maybe in a tape loop.
It sure can't sound right until you get the inverse network plus equalizer *flat*, so that's the place to start looking, as SY has said.
All good fortune,
Chris
It sure can't sound right until you get the inverse network plus equalizer *flat*, so that's the place to start looking, as SY has said.
All good fortune,
Chris
The groovewatt uses 12AX7 SRPP input driving the RIAA comp section feeding another 12AX7 SRPP stage followed by the 12AT7 cathode follower. I'm using this design with a home brewed PCB for my next project. I've got a set of 12AD7's along with several 12AX7's flavors to try out for tube rolling.
The problem is that you've chosen a really difficult and finicky project for your first time out of the gate. Low noise and high dynamic range are critical in this application, and that means great attention to detail in the design and layout of the circuit. So if you want a tube phono stage, you can either work up to it by doing some easier projects first, or build a design that's been debugged, proven, and well-supported.
Hi!
If you are serious about getting into DIY and want to learn some things, don't dismiss this project so quickly. Although it has it's flaws as some pointed out, it can still be made to work. Will probably not be the best phono ever, but surely can make a good learning piece. Since you already built it, why don't you get more into it, undestand it and learn enough from it to make your next one better.
The RIAA networks seem in the correct region, they probably did not take all the stray parameters into account, but surely are not way off.
I'd get a iRAA and measure it, then see were it deviates from the ideal and try to correct the RIAA. Most likely the 330pF needs to be lowered to give more 'air' at the top end. I'd also bypass the cathode resistors with 47 or 100uF.
The output is already unecessarily loaded with a low 22k. I'd increase that to 220k to ease the load. Even 1M is good there..
Important when you want to get into DIY is not to easily get discouraged. And get some basic measurement equipment. DVM, signal generator and scope are essential tools.
Best regards
Thomas
If you are serious about getting into DIY and want to learn some things, don't dismiss this project so quickly. Although it has it's flaws as some pointed out, it can still be made to work. Will probably not be the best phono ever, but surely can make a good learning piece. Since you already built it, why don't you get more into it, undestand it and learn enough from it to make your next one better.
The RIAA networks seem in the correct region, they probably did not take all the stray parameters into account, but surely are not way off.
I'd get a iRAA and measure it, then see were it deviates from the ideal and try to correct the RIAA. Most likely the 330pF needs to be lowered to give more 'air' at the top end. I'd also bypass the cathode resistors with 47 or 100uF.
The output is already unecessarily loaded with a low 22k. I'd increase that to 220k to ease the load. Even 1M is good there..
Important when you want to get into DIY is not to easily get discouraged. And get some basic measurement equipment. DVM, signal generator and scope are essential tools.
Best regards
Thomas
In the schematic you posted, the cathode bias resistors have been altered, why?
In the original circuit, were they capacitively decoupled?
Have in mind that the plate resistance from tube (Rp) is added to (mu+1)RK to form the definitively output resistance of the entire stage, and it then modifies time constants, given fixed external resistors and capacitors. So, the zeros and poles (Time constants) of RIAA are out of place in the freq/amplitude response chart of your amplifier.
Good luck!
In the original circuit, were they capacitively decoupled?
Have in mind that the plate resistance from tube (Rp) is added to (mu+1)RK to form the definitively output resistance of the entire stage, and it then modifies time constants, given fixed external resistors and capacitors. So, the zeros and poles (Time constants) of RIAA are out of place in the freq/amplitude response chart of your amplifier.
Good luck!
Last edited:
Hi Thomas,
You're right !!!! I've bypassed the cathode's with 100UF caps
And the 330PF reduced to 220pF (I had no other values )
and the output load increased to 220k , It is much better already !!!!
I have all the measuring tools, but not the knowledge to solve problems.
What is the best way to measure the riaa network , i have made a reverse riaa network but I can not verify the accuracy
You're right !!!! I've bypassed the cathode's with 100UF caps
And the 330PF reduced to 220pF (I had no other values
)and the output load increased to 220k , It is much better already !!!!
I have all the measuring tools, but not the knowledge to solve problems.
What is the best way to measure the riaa network , i have made a reverse riaa network but I can not verify the accuracy
Attachments
Last edited:
Once again. The output impedance seen since out of the tube (Including Ra or Rp), enter in the times constants RIAA needs. The cathode circuit of any valve (Triode, pentode, etc.) also indirectly makes this Rp. Suppouse a triode, your case. When the cathode resistor is UNBYPASSED, it reflects to the plate circuit as mu+1 bigger than the resistor itself. Example: say rk= 1k, mu = 20 Ra = 100K, umbypassed resistor is seen in the anode as:
RP = 100K + (20 +1)1K = 121Kohms.
In the example, bypassing the resistor, becomes rp only, then RP = 100K Then the output impedance is 21% more when unbypassed. The same result happen if you change the resistor value. Let reduce 1K to 500R:
RP = 100K + (20+1) 500R = 110.5K
The final value of output node is this RP in parallel with R load (The r from plate to +B).
Ergo, modifying ANY in the circuit, including plate resistor values DOES modify time constants, the RIAA response WILL be modified. Then, the audio response will be lost.
Pretty fine the photos!!!.
Good luck!!
RP = 100K + (20 +1)1K = 121Kohms.
In the example, bypassing the resistor, becomes rp only, then RP = 100K Then the output impedance is 21% more when unbypassed. The same result happen if you change the resistor value. Let reduce 1K to 500R:
RP = 100K + (20+1) 500R = 110.5K
The final value of output node is this RP in parallel with R load (The r from plate to +B).
Ergo, modifying ANY in the circuit, including plate resistor values DOES modify time constants, the RIAA response WILL be modified. Then, the audio response will be lost.
Pretty fine the photos!!!.
Good luck!!
Hi Martin,
nice that you could obtain some improvement already!
As for the iRIAA:
Use low tolerance (1%) parts to build it. You can verify it by running a frequency sweep and compare the output against a RIAA frequency response table. Or verify it with a known to be good phonostage. It should provide a linear frequency response if you feed it from a signal generator through the iRIAA.
Best regards
Thomas
nice that you could obtain some improvement already!
As for the iRIAA:
Use low tolerance (1%) parts to build it. You can verify it by running a frequency sweep and compare the output against a RIAA frequency response table. Or verify it with a known to be good phonostage. It should provide a linear frequency response if you feed it from a signal generator through the iRIAA.
Best regards
Thomas
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.