John in an earlier post suggested a way to reduce the effect of cable capacitance when the gain pot is located off-board: viewtopic.php?f=7&t=753&start=53
I tested a driven shield approach and drew up the schematic.
Flat Phono Preamp (Instrumentation Preamp) with Variable Gain Using Driven Shields to Reduce the Effect of Cable Capacitance.
The summing nodes at the inverting inputs of IC1 are very, very critical.
Any minute stray capacitance imbalance from either inverting input to ground is disastrous to high frequency common mode rejection.
Capacitance appearing across Rg (which includes those referenced to ground) can also cause HF response peaking.
See: That Corporation's "Analog Secrets Your Mother Never Told You" pdf pages 16-18 http://www.thatcorp.com/datashts/Analog ... ld_You.pdf
Avoid stray capacitance.
Flat Phono Preamp (Instrumentation Preamp). Avoid Stray Capacitance to Ground in the Gain Leads.
Because of cable capacitance a shielded cable with a grounded screen cannot be used for remote gain leads to a front panel pot.
When the cable shields are driven by the instrumentation amp outputs the effect of the unwanted capacitance is that it appears harmlessly in parallel with R8 and R9.
The effectiveness of the shields are the same as if they were grounded due to the low source impedance of the op amp outputs driving them.
Any small capacitance across Rg can also cause HF response peaking.
To avoid a shunt capacitance from appearing across Rg two separately shielded cables are required.
DO NOT use two conductor shielded cable and share a shield: Each shield needs to be driven by its respective output.
When constructing a unit with remote gain control its still important to keep the lead length as short as possible and the entire unit shielded in a metal enclosure.
Tests show that with careful lead dress using the approach shown in the schematic common mode rejection and low noise can be maintained.
My recommendation is that the gain pot be a stepped switch tapered reverse log.
The gain equation for the instrumentation amp is G = 1+[(R8+R9)/Rg]
Note that if THAT1240s are used after IC1 they provide an additional 6 dB gain.
Phono Transfer System Construction Information
- mediatechnology
- Posts: 5444
- Joined: Sat Aug 11, 2007 2:34 pm
- Location: Oak Cliff, Texas
- Contact:
Good digital RIAA EQs ... accurate and Minimum Phase
Wayne, I think you posted somewhere a short list of the digital RIAA EQs which you'd found good ... accurate and Minimum Phase ... and the ones which weren't.
Could you please help a beach bum with the info
I've searched the 2 huge phono preamp threads and this one without success
Could you please help a beach bum with the info
I've searched the 2 huge phono preamp threads and this one without success
- mediatechnology
- Posts: 5444
- Joined: Sat Aug 11, 2007 2:34 pm
- Location: Oak Cliff, Texas
- Contact:
Re: Phono Transfer System
Were you looking for .vst?
These are based on the DH Biquad from Jim W: viewtopic.php?f=15&t=782
And these from our forum member Olaf Matthes: viewtopic.php?f=15&t=784
The built-in ones for Diamond Cut 8.5 are very good: http://www.diamondcut.com/store/index.p ... age=page_2
On my last test I used Olaf's and I casually compared them to analog. They sounded pretty close.
Looks like a good test for AudioDiffmaker.
I've also used the DH Biquads but for some reason couldn't get them to run correctly in one installation of Audacity and haven't bothered to troubleshoot it.
These are based on the DH Biquad from Jim W: viewtopic.php?f=15&t=782
And these from our forum member Olaf Matthes: viewtopic.php?f=15&t=784
The built-in ones for Diamond Cut 8.5 are very good: http://www.diamondcut.com/store/index.p ... age=page_2
On my last test I used Olaf's and I casually compared them to analog. They sounded pretty close.
Looks like a good test for AudioDiffmaker.
I've also used the DH Biquads but for some reason couldn't get them to run correctly in one installation of Audacity and haven't bothered to troubleshoot it.
- mediatechnology
- Posts: 5444
- Joined: Sat Aug 11, 2007 2:34 pm
- Location: Oak Cliff, Texas
- Contact:
The Case for Balanced Phono Cartridge Wiring
I'm about to do some Analog/DSP RIAA EQ comparisons and before I did wanted to grab a screenshot of the RIAA-EQ'd noise floor.
The first pic is the unweighted noise floor with both channel's inputs balanced.
The cartridge connections are shielded twisted pair.
The second pic is with the Left channel jumpered for unbalanced operation. No common mode rejection in the cartridge signal lines is realized in the Left channel.
Flat Moving Magnet Phono Preamp Noise Floor Both Channels Balanced.
Flat Moving Magnet Phono Preamp Noise Floor Left Channel (Blue) Is Unbalanced.
They may measure almost identically but they do sound different.
Update: December 18, 2018:
The original post at AudioScience read:
The first pic is the unweighted noise floor with both channel's inputs balanced.
The cartridge connections are shielded twisted pair.
The second pic is with the Left channel jumpered for unbalanced operation. No common mode rejection in the cartridge signal lines is realized in the Left channel.
Flat Moving Magnet Phono Preamp Noise Floor Both Channels Balanced.
Flat Moving Magnet Phono Preamp Noise Floor Left Channel (Blue) Is Unbalanced.
They may measure almost identically but they do sound different.
Update: December 18, 2018:
I'm talking about the signature of the noise floor sounding different. The noise level measurements, in dB, are nearly-identical, but the FFT, and the audible noise floor signature, are quite different in the inductive region of the cartridge above 1 kHz.svart-hvitt: Almost identically means balanced measures a little better. Maybe the sonic differences are due to factors not covered by the measurements? Or maybe he hears voices in the air? https://www.audiosciencereview.com/foru ... post-49516
The original post at AudioScience read:
It is complete BS. The balanced, flat, preamp in this thread, when they're available, is a $39 kit. The lowest-noise performance is obtained using a 30 cent IC.And, yet, we have phono stages costing thousands of dollars, with big toroidal power supplies that reek of engineering overkill for such small signals, and claims of new innovative topologies for a circuit that should have been solved 30+ years ago.
Why?
Can one of you EE or other electronics types explain why all this is needed?
It just seems like complete BS to me.
Re: Phono Transfer System
Hello, finally getting my build completed, Will be using as stand alone to feed a pcm4204evm adc. My phono leads are true balanced as shown in this thread though so I wonder, do I need to populate c4-c5 c-24 c-25? What value and voltage for those caps? Thanks!
- mediatechnology
- Posts: 5444
- Joined: Sat Aug 11, 2007 2:34 pm
- Location: Oak Cliff, Texas
- Contact:
Re: Phono Transfer System
Thank you for joining us quattro64!
I have never used the output coupling capacitors since the output offset is usually quite low with a balanced cart connection.
I had to look at the PCM4202 input stage to see if it has Vcm present on the inputs.
It does.
The PCM4202EVM uses an OPA1632 with Vcm supplied by the 4202.
Without input coupling capacitors installed on the EVM it does have DC just like my PCM4222 EVM. (C1-C4 on the 'EVM).
See: viewtopic.php?f=6&t=886
If the Flat preamp is going to be driving the PCM4202EVM directly you might want to install C4, C5, C24 and C25 with the + terminals pointing to the PCM4202EVM or install C1-C4 on the 'EVM.
Otherwise the THAT1240 outputs will have to sink about 2.5 mA DC.
Driving my PCM4222 EVM directly-coupled with most op amp outputs I don't see any increase in distortion but when you patch or switch the inputs it is pretty clicky since there's DC on the A/D inputs.
I have never used the output coupling capacitors since the output offset is usually quite low with a balanced cart connection.
I had to look at the PCM4202 input stage to see if it has Vcm present on the inputs.
It does.
The PCM4202EVM uses an OPA1632 with Vcm supplied by the 4202.
Without input coupling capacitors installed on the EVM it does have DC just like my PCM4222 EVM. (C1-C4 on the 'EVM).
See: viewtopic.php?f=6&t=886
If the Flat preamp is going to be driving the PCM4202EVM directly you might want to install C4, C5, C24 and C25 with the + terminals pointing to the PCM4202EVM or install C1-C4 on the 'EVM.
Otherwise the THAT1240 outputs will have to sink about 2.5 mA DC.
Driving my PCM4222 EVM directly-coupled with most op amp outputs I don't see any increase in distortion but when you patch or switch the inputs it is pretty clicky since there's DC on the A/D inputs.
Re: Phono Transfer System
Thanks!, I'm glad you are familiar with these EVMs. I looked at the pcm4204evm guide and the schematic shows the currently blank spots for the possible caps. It looks like it wants electrolytics. Could film caps work? This is for dc blocking, correct? Schema also shows an R in parallel with the cap. Both have undetermined values?? What vdc and uF should I go for? I would prefer to use caps on the evm and then jump the connections on the MM phono. Thanks again.
- mediatechnology
- Posts: 5444
- Joined: Sat Aug 11, 2007 2:34 pm
- Location: Oak Cliff, Texas
- Contact:
Re: Phono Transfer System
The PCN4202 EVM does want electrolytics for DC blocking.
I agree that it would be best to add the caps to the EVM.
The input impedance is low - around 1K - so you would need at least 47 uF.
The parallel resistors jumper the caps when they are not installed so you will need to remove them.
A 25V cap should be sufficient.
I agree that it would be best to add the caps to the EVM.
The input impedance is low - around 1K - so you would need at least 47 uF.
The parallel resistors jumper the caps when they are not installed so you will need to remove them.
A 25V cap should be sufficient.
Re: Phono Transfer System
Thanks, 1k is low so yes I will start with 47uf, impedance mismatch shows up in either rolled off highs or no bass, I'll experiment if 47uf doesn't work. I have a bunch of 47uf 16v, will those work? So the plan is to jump the c4-c5, c24-c25 on the MM then remove any jumpers or Rs on the evm and replace with polar lytics, correct?