
Home  Forums  Rules  Articles  diyAudio Store  Blogs  Gallery  Wiki  Register  Donations  FAQ  Calendar  Search  Today's Posts  Mark Forums Read  Search 
Analogue Source Turntables, Tonearms, Cartridges, Phono Stages, Tuners, Tape Recorders, etc. 

Please consider donating to help us continue to serve you.
Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving 

Thread Tools  Search this Thread 
16th February 2019, 08:51 PM  #481 
diyAudio Member
Join Date: Mar 2005
Location: Blaricum

Hi Jack,
Allow me to give you a simple tip. With 50 Ohm for R19, the equivalent input noise at R19 for a 10Hz filter @3Khz and with the OPA2604 will be 38nV. For a SNR of 35dB to give an accurate measuring result, the Signal will then have to be 2.1uV at R19. With this 50 Ohm resistor, voltage over the Cart will then be 400uV. When replacing R19 by a 5K resistor, equivalent noise wil be 46nV with a 10Hz filter. For the same 35dB SNR you will need 2.6uV signal at this spot. However signal over the cart will now be 5 uV, or 80 times lower as with the 50 Ohm resistor. Your 3577 will calculate a wrong Inductance for the coil, that only has to be corrrected with a factor 100. Hans 
17th February 2019, 04:06 AM  #482 
diyAudio Member
Join Date: Dec 2005
Location: Kuala Lumpur

I have got usable results, showing no impedance or phase changes at all with the source at 50dBU, which means 80dBU at the sense resistor

17th February 2019, 08:28 AM  #483  
diyAudio Member
Join Date: Mar 2005
Location: Blaricum

Quote:
Great results so far, already down at ca 60uV on the Cart. If possible, could you show the same recording as in #470, but now for the 80dBu at the sense resistor recording. I’m especially interested in how much more noise there is on the curves compared to the earlier recording at 40dBU. Hans 

17th February 2019, 08:45 AM  #484 
diyAudio Member
Join Date: Mar 2005
Location: Blaricum


17th February 2019, 11:07 AM  #485  
diyAudio Member
Join Date: Dec 2005
Location: Kuala Lumpur

Quote:
Again impedance is scaled by 1/5, so the maximum value is really 50k I used 2M samples, 4M gets an out of memory error on a 8GB PC The TLV2372 is a noisy opamp, the only rail to rail I had to hand. Also a metal box would be a good idea. 

17th February 2019, 05:02 PM  #486  
diyAudio Member
Join Date: Mar 2005
Location: Blaricum

Quote:
Impedance is calculated as: Rsense*V(r)/(V(a)V(r)). V(a) and V(r) both being contaminated with noise of a certain percentage. When V(a) and V(r) in the denominator are coming close, the error in the calculation of Z becomes bigger and bigger. That's what I think that can be seen in your image. The higher the frequency, the closer V(a) and V(r) become and the higher the noise in the calculation. Say V(a) is 100 and V(r) is 95, both carrying +/1% noise. V(r)/(V(a)V(r)) may now vary between 94/(10194) and 96(9996), or between 658 and 288. The original error of 1% has now become +/39%. The only remedy to keep the voltages apart, is to increase Rsense. This won't increase the noise, because your Cmos amps will stay to be the dominant noise contributors. So my question is to prove the theory and to get a more accurate measurement, could you repeat the measurement with Rsense 20K or even 50K. Hans 

Yesterday, 12:16 AM  #487 
diyAudio Member

Thanks Hans.
Am considering using the ADA4817 which is a single with 4nV/Rt Hz noise, and using the headers to adjust the gain via parallel resistors. ADA4817 is very high bandwidth so needs particular attention to layout. Yes, the HP3577 is not happy with DC on the input. The Bode 100 is more tolerant. 
Today, 08:16 AM  #488 
diyAudio Moderator

Thanks David.
As you’ve said, your 0dBu is 0.75Vrms. Accordingly, –80dBu is 75uV. For the G1025 (Sensitivity 6.5 mV +/1.5 dB, 1 kHz @ 5 cm/sec) this corresponds to a signal level close to 40dB below the recording reference level of 5cm/sec. This is adequately low, no need IMO for going lower. George
__________________
["Second Law is a bitch."  SY] ["I insist on respecting the means of the average person working in their garage/basement." Scott Wurcer] 
Today, 10:27 AM  #489 
diyAudio Member
Join Date: Mar 2005
Location: Blaricum

@ LD,
It was you who came with the idea to investigate whether inductance is dependant on signal level for an MM Cart. Question is now, has this been answerred to your satisfaction with proper accuracy and low enough level ? Hans 
Thread Tools  Search this Thread 


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
PTC Behaviour  jayam000  Solid State  11  4th May 2009 04:35 AM 
F3  LED behaviour  Nicola  Pass Labs  2  22nd August 2007 09:02 AM 
New To Site?  Need Help? 