"WHAMMY" Pass DIY headphone amp guide

I
The AD825 is a very hi speed device and requires care in its application and use. I assume you used an adapter board and two AD825's as they are single opamps.

I've always rated the TL0 series very highly and still do. The LM833 is another good device but one that is available in two versions, the original LM833-N from TI and also a TI version with a quasi complementary output stage called LM833.
I have the lm833-N, is it right?
 
This might sound heretical / old-fashioned / retrograde, here goes: with proper bypassing the NE5532 works just fine for 90% of use cases. We're talking 22 to 47 ohm resistor on each rail, followed by about 220 uF electrolytics in parallel with 3 to 10 uF X7R ceramics before pins 4 and 8 [1]. I do run into small clicks when following a Khozmo switched attenuator elsewhere in the preamp, probably due to input bias, so my next trick will be an OPA2134 or other FET input opamp after the control.

[1] A post on reasons why will happen eventually. Wayne Colburn demonstrated this topology to great effect in the WHAMMY, sans ceramics, and the results speak for themselves.
 
Only the Texas Instruments (TI) version is available with the Quasi stage.

I have the lm833-N, is it right?

If it is Texas brand it is the old version. Any old stock National Semiconductor versions are all the non Quasi type.

This is the reply I received many years ago from TI.

LM833-N is an op-amp originally from National Semiconductor. NS named this device as LM833. As for the package, LM833N means a PDIP package LM833-N. NS calls PDIP package as “N” package.

TI developed another LM833 as 2nd source device and named it LM833. When TI made LM833, TI changed its output stage to NPN type from VPNP.

After TI acquired NS, TI renamed NS’s LM833 to LM833-N in order to differentiate the LM833-N from TI’s LM833. Please don’t be confused by LM833-N and LM833N.

This is the Quasi:

Screenshot 2024-07-09 052836.png


This is the original:

Screenshot 2024-07-09 053021.png
 

Attachments

  • Like
Reactions: Jaccher
NOPB simply means no lead (lead free). Either type will work and the original was still very highly regarded. LM833N was the original NPN/PNP outputs stage.

This is something I posted a few years ago. Original vs Quasi LM833 driving a 100 ohm load. Scope on 5v/div. The WHAMMY won't come close to needing this drive ability.

Screenshot 2024-07-09 074226.png
 
  • Like
Reactions: Mark Johnson
@Jaccher , Muses 02 opamp sounds really good. It does not oscillate but needs 68pF compensation to get the sqares right.
When I did the Whammy, I ran out of OPA1656, so I couldn't determine if it oscillates and if I need some compensation for that opamp. That should be a good option as well, however, you never know without trying.
 
If you like TL072, then Muses 02 is the right choice for you + some quality 68pF (2pcs) for step response compensation. And the OPA2604 has that sound character too, but it sounds better than the TL072 or OPA2134. It's something in between, and doesn't need compensation. OPA2604 is an obsolete opamp. Max. voltage supply for Muses 02 is +-16V DC.
 
Last edited:
Yes, procurement is problematic. I have them from long ago, when there were no copies, only BB. The OPA 2134 is like a replacement for the OPA2604, but that OPA never sounded good to me, not even close to the 2604.

PS

I bought a few used OPA2604 on ebay, DIP8 in ceramic case as well as SOIC 8 SMD version. I can't find any difference from the old originals in the DIP8 plastic case. Neither by measuring nor listening.

I have a new plastic fake OPA2604, purchased at a local store. It works of course but it sounds just like a TL072, and there is probably some kind of TL072 inside.
 
Last edited:
  • Like
Reactions: Mooly
I would have thought any fakes were more than likely something like a 4558 tbh.

A check of the supply current is the first basic check to see if something is fake followed by a quick look at its bandwidth and then whether it seems to be a FET or bjt input. Also look at its output current capability. A 2604 will fully drive 600 ohm loads while a T072 will fall well short of doing that.
 
Who knows what they put in those copies? Anyway, I used fake NE5532, TL072 and OPA2604 in DC protection, DC servo and voltage regulators for non-audio purposes. Not a single one has died so far. For audio, they are not good at all, they sound bad. I mostly use them for rehearsals when I'm making something. If everything is fine, I put more expensive opamps. I still have to try the OPA1656, I'm sure it sounds better than the 2604, but the question is whether it oscillates and needs compensation in Whammy circuit.
 
Try it, it's cheap. If seller doesn't lie that they were removed from old devices, then they are original.

Check the DC offset before connecting the headphones, it should be small 1-2mV. Also, measure the DC offset when you turn off the amp. Measure for a longer time and you will see what happens.
 
  • Like
Reactions: Jaccher