Cmoy vs Chipamps

Status
Not open for further replies.
I have a lm4880 headphone amplifier and I want to know if a cmoy with an opamp opa2134 will sound better than a lm4880 or a TPA152 chips.

Which one is the best amplifier, chipamp or cmoy opa2134?

Regards
Alfredo Mendiola
Lima, Peru
 
That's an apples to oranges comparison.
LM4880 and TPA152 are optimized for low-voltage single supply operation (2.7 - 5.5 V and 4.5 - 5.5 V, respectively) and can swing almost rail to rail on the output with light loads. They were designed to drive loads of 32 and even 8 ohms directly and have a corresponding amount of "oomph", though low idle current means they're not super-duper fast and show a fair bit of high-frequency (crossover?) distortion with such loads. Inverting operation is virtually a must as they would run into common-mode range limitations otherwise. That plus them being MOS parts also means they're more or less noisy (with a lot of 1/f in particular). You must use output coupling capacitors with these parts, which get kinda bulky with hi-fi requirements - I'd consider 220 µF the bare minimum for a 32 ohm load, 470 would be better. These presumably take up about as much space as the rest of the circuit.
On the upside, you could run an LM4880 based amp on 3 AA cells for approximately forever and a day - literally hundreds of hours. Not bad considering it's a rather usable headphone amp already. It obviously won't go arbitrarily loud into high-impedance cans of modest sensitivity.

While the OPA2134 will work down to 5 V (+/- 2.5 V) and its output will swing reasonably close to the rails, it is optimized for operation at higher voltages, up to +/-18 V, and driving loads in the hundreds of ohms or even kOhms to correspondingly high voltages. Vs. the more expensive '132, this model requires a bit more voltage still. It can supply a decent amount of output current, but load immunity in terms of high-frequency distortion is only so-so. (At least its idle current is rather moderate.) Capacitive load driving ought to be decent, but a bit of series resistance still is recommended.
This part probably is best suited for a dual 9V supply cMoy with a DC-coupled output driving mid- to high impedance headphones. Those typically don't mind a few dozen ohms worth of output impedance either. Sensitive dynamic driver in-ears should also work well though. Some attention to input impedance nonlinearity is required in non-inverting operation, so I'd use a 10k volume pot rather than the commonly-recommended 50k type.
I can see why this one would've been used for a cMoy rather than the trusty NE5532, whose common-mode distortion characteristics aren't too great though load immunity is rather better (an inverting circuit makes good use of this part and has been used in the BTech BT928 amplifier). If you need a serious amount of current and want to drive loads down to 32 ohms directly, however, the part to use is the NJM4556. It arguably has the most in common with some of the dedicated headphone drivers, though it's an older bipolar part that'll withstand operation at up to +/- 15 V (no more than +/- 9 or 12 recommended for low-impedance loads though). Here are some measurements for an amplifier using one. Its main downside is relatively high idle current consumption, twice that of the OPA2134 (8 mA vs. 4 mA). This part was originally developed for things like CD players in the mid-'80s, where power consumption wasn't such an extreme issue. Noise also is somewhat higher.

Obviously a 4556 based cMoy with two 9V, 200 mAh batteries would be expected to run for about 25 hours only. Rechargeables recommended.
 
Status
Not open for further replies.