Au contraire, about the only thing that's simple to investigate, and which has a remote possibility of explaining "works one way with JFET opamps, works a different way with BJT opamps" is: wrong resistor values around the input pins.
Other and possibly better diagnostic procedures exist, and I'm sure the membership will recommend them - - - - but those procedures require a lot more work than just "read the codes and check the schematic"
Other and possibly better diagnostic procedures exist, and I'm sure the membership will recommend them - - - - but those procedures require a lot more work than just "read the codes and check the schematic"
Well, I carefully did measure all the resistors and everything checks out. I popped in the NE5532 and it worked perfectly, offset readings of about -4.5mV and -5.5mV. What the heck?
I put the LME49720 back in again. Offset back at -115mV on the left channel. Absolutely baffled. Then I wiggled the chip around in the socket...
I must sheepishly admit the offset issue was caused by an iffy connection at the op amp socket. Oops. Down to less than +/- 0.5mV with the LME. Now we're talkin'!
Here's the schematic with the values I used highlighted in red. Nothing too wacky, a couple values were chosen because they are convenient, but most were chosen because they keep the offset to within a couple millivolts while accounting for the NE5532's input bias current. Plus, I wanted a gain of exactly 4, again for convenience (I always set every preamp I build for 12dB of gain so I can swap them around and keep the gain structure constant). 47uF Nichicon bipolar Muse caps are very good to use for this circuit and push the -3dB point down to just a couple Hz. For the input caps, 2.2uF/100V WIMA MKP10s with the 27.5mm pitch fit perfectly and are probably as good as any Mundorf or Jantzen you can buy.
Note that I'm running the output devices at ~120mA. Not noted here is that I used IRF610 and IRF9610 FETs and green 3mm LED voltage refs for the regs to yield +/-17 V. And I put a 0.1uF poly cap across the positive and negative power pins of the op amp. One more tip: use a thermistor at the AC inlet, and consider larger resistors (10ish ohms) in the power supply. These will both save your transformer and regs a bit. As it is, I might add bigger resistors to the the PSU, as the regs still see +/-26ish raw volts and run HOT.
More pics:
Made with a Modushop chassis, the same as the Whammy kit, but with the 3mm front plate. I realized that if you are OK with a captive cord and screw-in fuse compartment, you don't have to cut a weird-shaped hole for an IEC inlet. Joke's on me, I didn't have the right tools. Spent a lot of time with a metal file getting those holes right... Just buy the kit from the store.
I do have source switching capability and two line-level outputs, though neither is hooked up yet. The third switch, closest to the output jack, is an auxiliary so I can add a crossfeed or bass boost circuit later.
I'm listening with the LME49720 in place right now, perfectly content... This is probably one of the very best sounding headphone amps out there without having to spend thousands. Thanks Wayne!!
I put the LME49720 back in again. Offset back at -115mV on the left channel. Absolutely baffled. Then I wiggled the chip around in the socket...
I must sheepishly admit the offset issue was caused by an iffy connection at the op amp socket. Oops. Down to less than +/- 0.5mV with the LME. Now we're talkin'!
Here's the schematic with the values I used highlighted in red. Nothing too wacky, a couple values were chosen because they are convenient, but most were chosen because they keep the offset to within a couple millivolts while accounting for the NE5532's input bias current. Plus, I wanted a gain of exactly 4, again for convenience (I always set every preamp I build for 12dB of gain so I can swap them around and keep the gain structure constant). 47uF Nichicon bipolar Muse caps are very good to use for this circuit and push the -3dB point down to just a couple Hz. For the input caps, 2.2uF/100V WIMA MKP10s with the 27.5mm pitch fit perfectly and are probably as good as any Mundorf or Jantzen you can buy.
Note that I'm running the output devices at ~120mA. Not noted here is that I used IRF610 and IRF9610 FETs and green 3mm LED voltage refs for the regs to yield +/-17 V. And I put a 0.1uF poly cap across the positive and negative power pins of the op amp. One more tip: use a thermistor at the AC inlet, and consider larger resistors (10ish ohms) in the power supply. These will both save your transformer and regs a bit. As it is, I might add bigger resistors to the the PSU, as the regs still see +/-26ish raw volts and run HOT.
More pics:
Made with a Modushop chassis, the same as the Whammy kit, but with the 3mm front plate. I realized that if you are OK with a captive cord and screw-in fuse compartment, you don't have to cut a weird-shaped hole for an IEC inlet. Joke's on me, I didn't have the right tools. Spent a lot of time with a metal file getting those holes right... Just buy the kit from the store.
I do have source switching capability and two line-level outputs, though neither is hooked up yet. The third switch, closest to the output jack, is an auxiliary so I can add a crossfeed or bass boost circuit later.
I'm listening with the LME49720 in place right now, perfectly content... This is probably one of the very best sounding headphone amps out there without having to spend thousands. Thanks Wayne!!
Last edited:
Hi all,
I'd like to report that i get an excellent synergy between my whammy with a V6 classic Op Amp and the Hifiman HE1000 Stealth V2.
The relatively wider soundstage and more dynamic / strident sound of the V6 Vivid did not shine through here, while it did on the HD58X.
Instead, the V6 classic's pristine transients, natural tone and organic sound really pairs well with the highly resolving and great soundstage of HE1000 Stealth V2.
Bass / low mids could be a bit cleaner, I may want to try the V7 Classics on a new whammy build.
Did anyone compare the whammy to the CFA3 amp?
I'd like to report that i get an excellent synergy between my whammy with a V6 classic Op Amp and the Hifiman HE1000 Stealth V2.
The relatively wider soundstage and more dynamic / strident sound of the V6 Vivid did not shine through here, while it did on the HD58X.
Instead, the V6 classic's pristine transients, natural tone and organic sound really pairs well with the highly resolving and great soundstage of HE1000 Stealth V2.
Bass / low mids could be a bit cleaner, I may want to try the V7 Classics on a new whammy build.
Did anyone compare the whammy to the CFA3 amp?
Last edited:
Goood morning!
Anyone measured SINAD on the whammy stock and "pimped up" versions? Just to see how it compares with newer offerings. I know it's just one measurement out of many, but 6L6 did provide good data on THD and noise in his write up which is very useful.
Anyone measured SINAD on the whammy stock and "pimped up" versions? Just to see how it compares with newer offerings. I know it's just one measurement out of many, but 6L6 did provide good data on THD and noise in his write up which is very useful.
Here's how the LM833EN measured on my whammy:
Here at highest level just before clipping on 150Ohms load, 9.6Vrms = 614mW, harmonics are >100dB below fundamental.
ASR-style SINAD would probably fare a bit lower with 4Vrms output..Without my mods I expect it to measure roughly the same.
But I can tell you that this amp sounds miles better than a Topping A30 pro which measured much better, so SINAD is far from depicting the whole performance!
Here at highest level just before clipping on 150Ohms load, 9.6Vrms = 614mW, harmonics are >100dB below fundamental.
ASR-style SINAD would probably fare a bit lower with 4Vrms output..Without my mods I expect it to measure roughly the same.
But I can tell you that this amp sounds miles better than a Topping A30 pro which measured much better, so SINAD is far from depicting the whole performance!
Yes, the harsh reality. 🙄
I don't know how someone hasn't already discovered some measurement (s) that is directly related to sound quality, that's strange to me. After so many years of audio technology development.
I don't know how someone hasn't already discovered some measurement (s) that is directly related to sound quality, that's strange to me. After so many years of audio technology development.
Ohh interesting, thanks for that. It does lag behind the newer offerings quite a bit, but im not complaining. I wonder if fet op amps would perform better?
Sound quality is very individual, so impossible to quantify using a single measurment. It's a bit like wine tasting!
Sound quality is very individual, so impossible to quantify using a single measurment. It's a bit like wine tasting!
I wonder which version that one is, whether the original with NPN/PNP output stage or the later NPN quasi.
Its super confusing which is which. This is what TI told me around 10 years ago.
So its either this one:
or this one:
The original John Linsley Hood rated these very highly and at one time said it 'might be the best yet'.
I have drawn this on LT SPICE, but i am getting no output just a negative offset, someone please can help me to figure out what actually going wrong.
Regards.
I just want to simulate it with OPA1656
Regards.
I just want to simulate it with OPA1656
Anyone here build the SMD version posted here?
https://www.diyaudio.com/community/threads/whammy-pass-diy-headphone-amp-guide.317803/post-6630881
It lists r15,r23,r28,r33 as 5Kohm instead of the 10kohm in the original schematic. It also lists r16,r22,r29,r32 as 1 ohm instead of 10 ohms in the original. My mosfets get super hot very quickly, I have only let it run for 30 seconds.
https://www.diyaudio.com/community/threads/whammy-pass-diy-headphone-amp-guide.317803/post-6630881
It lists r15,r23,r28,r33 as 5Kohm instead of the 10kohm in the original schematic. It also lists r16,r22,r29,r32 as 1 ohm instead of 10 ohms in the original. My mosfets get super hot very quickly, I have only let it run for 30 seconds.
Hi @gsrchrisu
I'm no expert but 1Ohm output resistor seems suspiciously low. The output Bias current is directly determined by this resistor value as the voltage across it is determined by the Optocoupler's LED.
I would think that it's a typo, and start with 10Ohms instead of 1Ohm as in original BOM.
I'm no expert but 1Ohm output resistor seems suspiciously low. The output Bias current is directly determined by this resistor value as the voltage across it is determined by the Optocoupler's LED.
I would think that it's a typo, and start with 10Ohms instead of 1Ohm as in original BOM.
Hi guys,
Has anyone tried a Sparkos SS3602 dual discrete op amp in The WHAMMY? If so, what is your assessment? I currently have a JRC 8102A in place. It sounds great, but you always wonder.....would it sound even better with discrete components....
Thanks,
John
Has anyone tried a Sparkos SS3602 dual discrete op amp in The WHAMMY? If so, what is your assessment? I currently have a JRC 8102A in place. It sounds great, but you always wonder.....would it sound even better with discrete components....
Thanks,
John
I tried it, it sounded quite sharp and oscillated. It is necessary to put 330pF compensation capacitors. In the end I adopted Muses 02 with 68pF compensation caps. Much calmer sound in the highs.