I tried using the OP27EZ in place of the OPA627 in one channel last night. Nulling the DC offset in this chip seems to be a 'good idea'.
Thankfully, there is a set of three pads that can have their traces cut that can be used to mount the 100K trim pot I had (10K optimal). The OP27 is all N-channel BJT and specs out quieter than the OPA627 FET based op amp.
I prefer this chip to the OPA627; it has a quieter background and it seems to reproduce longer reverb/decay than the 627. Generally sounds cleaner and less congested than the stock circuit. I will be mounting the other trimmer tonight and report my findings.
Beware: the offset adjust pins are different on the OPA627 than on the OP27. Once this circuit is added, the parts are no longer compatible.
Thankfully, there is a set of three pads that can have their traces cut that can be used to mount the 100K trim pot I had (10K optimal). The OP27 is all N-channel BJT and specs out quieter than the OPA627 FET based op amp.
I prefer this chip to the OPA627; it has a quieter background and it seems to reproduce longer reverb/decay than the 627. Generally sounds cleaner and less congested than the stock circuit. I will be mounting the other trimmer tonight and report my findings.
Beware: the offset adjust pins are different on the OPA627 than on the OP27. Once this circuit is added, the parts are no longer compatible.
Last edited:
The slew rate on the op27 is 2.8v/us verses the slew rate of the op637 being a minimum of 100v/us. It is easier to fabricate a low noise component when sacrificing slew rate speed. However the transient information of the music will also be sacrificed. Nothing inherently wrong with these trade-offs. Just understand, what might be, unintended consiquences.
Most real music has a slew rate under that of 2.8 V/uS. The only excepetion I am aware of is the rim shot. It is because of the low slew rates on the OP27 that putting them in the shunt regulator without swapping out the op637 is a "bad idea". That actually sounds awful.
I've found no sonic issue with using them in the first stage with no other changes.
I am sure that it would be better to use a faster and quieter opamp, but there is so little room on that board to attach the bypass caps necessary to run something like an LME499990. but truthfully, I've not tried attaching the bypass caps to the pins under the boards....
Also, these were just opamps that I had lying around. Thankfully, I like them a lot.
I've found no sonic issue with using them in the first stage with no other changes.
I am sure that it would be better to use a faster and quieter opamp, but there is so little room on that board to attach the bypass caps necessary to run something like an LME499990. but truthfully, I've not tried attaching the bypass caps to the pins under the boards....
Also, these were just opamps that I had lying around. Thankfully, I like them a lot.
For those of you that want more bass, try changing C2 to 0.0047uF - 0.0056uF. That's 4700pF - 5600 pf.
I will be doing that once I find a good cap and report back -- 5600 gives a lower sum of squares when calculating deviation from RIAA.
Also, replace 10uF C6 with 47uF is good too. 22-47 should be fine.
These figures yeild best numbers in my spreadsheet.
I will be doing that once I find a good cap and report back -- 5600 gives a lower sum of squares when calculating deviation from RIAA.
Also, replace 10uF C6 with 47uF is good too. 22-47 should be fine.
These figures yeild best numbers in my spreadsheet.
Last edited:
In my spreadsheet, I've changed C2=56nF, C6=47uF, and both 330K resistors to 300K.
This decreased the sum of squares of the error from RIAA using 78 frequency samples from 220 to 17.
My spreadsheet is not really ready for posting, but I would be happy to email it to you.
What software is that you used?
This decreased the sum of squares of the error from RIAA using 78 frequency samples from 220 to 17.
My spreadsheet is not really ready for posting, but I would be happy to email it to you.
What software is that you used?
Last edited:
I simulate this in LTSpice.
Calculator RC: RC pad corner frequency upper and lower cutoff frequency calculation filter calculate time constant tau RC voltage power calculator capacitance resistance - sengpielaudio Sengpiel Berlin
1620R+56n >> 1754Hz instead of 2122Hz
1620R+47n >> 2090Hz (oryginal)
1596R+47n >> 2121,73Hz - best.
Balanced input - I think this should work: https://picasaweb.google.com/115453...hkey=Gv1sRgCL7LrsLfpY3EaA#5957943693590460690
But you have to be sure about wiring in your tonearm.
Calculator RC: RC pad corner frequency upper and lower cutoff frequency calculation filter calculate time constant tau RC voltage power calculator capacitance resistance - sengpielaudio Sengpiel Berlin
1620R+56n >> 1754Hz instead of 2122Hz
1620R+47n >> 2090Hz (oryginal)
1596R+47n >> 2121,73Hz - best.
Balanced input - I think this should work: https://picasaweb.google.com/115453...hkey=Gv1sRgCL7LrsLfpY3EaA#5957943693590460690
But you have to be sure about wiring in your tonearm.
Thanks!
You're looking at the first stage. I'm not suggesting replacing 47n with 56n, I am suggesting replacing 0.01uF with 56n in the second stage, along with 10uF -> 47 uF and both 330K -> 300K.
The spreadsheet does:
NPDgain(Hz) = PDgain(Hz)) - PDgain(1000)) in db, where PDgain is the gain of the PD phono stage. Then calculate NRIAA(Hz) = RIAAgain(HZ) - RIAAgain(1000) in db.
Then errors are calculated as sum( (NPD(Hz) - NRIAA(Hz))^2).
I will make some measurements tonight. First step, using Jim Hagerman's Inverse RIAA hooked up to the phono stage.
Parts will be ordered today to make the changes.
You're looking at the first stage. I'm not suggesting replacing 47n with 56n, I am suggesting replacing 0.01uF with 56n in the second stage, along with 10uF -> 47 uF and both 330K -> 300K.
The spreadsheet does:
NPDgain(Hz) = PDgain(Hz)) - PDgain(1000)) in db, where PDgain is the gain of the PD phono stage. Then calculate NRIAA(Hz) = RIAAgain(HZ) - RIAAgain(1000) in db.
Then errors are calculated as sum( (NPD(Hz) - NRIAA(Hz))^2).
I will make some measurements tonight. First step, using Jim Hagerman's Inverse RIAA hooked up to the phono stage.
Parts will be ordered today to make the changes.
ahaja,
I see my mistake.... not 56nF but 5.6nF... sorry. (0.0056uF)
Could you also change the 10uF to 47uF please?
Thanks!
My spreadsheet says these are to be changed from design:
C2 = 05.6nF ( was 10nF) , C3 = 47uF (was 10uF), R4=R6=300K (was 330K).
Thanks for your help, and sorry for the confusion!
I see my mistake.... not 56nF but 5.6nF... sorry. (0.0056uF)
Could you also change the 10uF to 47uF please?
Thanks!
My spreadsheet says these are to be changed from design:
C2 = 05.6nF ( was 10nF) , C3 = 47uF (was 10uF), R4=R6=300K (was 330K).
Thanks for your help, and sorry for the confusion!
Last edited:
The link shows "Internal server error" will check later.
I tried taking some measurements, but I was a bit uncertain of them -- With the Jim Hagermann inverse RIAA, set at -60 db, I get some very consistent numbers on the stock unit.
However, the output does vary a bit with frequency, in the way that my spreadsheet suggests. I only have a meter that goes into the millivolts, and an analog signal generator, so it is difficult to make direct measurement.
I tried taking some measurements, but I was a bit uncertain of them -- With the Jim Hagermann inverse RIAA, set at -60 db, I get some very consistent numbers on the stock unit.
However, the output does vary a bit with frequency, in the way that my spreadsheet suggests. I only have a meter that goes into the millivolts, and an analog signal generator, so it is difficult to make direct measurement.