LME49740 vs TL074, design considerations and stability

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I recently spent an afternoon troubleshooting my sub filter... turns out the LME49740 would need output resistors to be stable, is that correct?
Yes, of course the power supplies are properly decoupled with 150 nF directly on each chips supply pins.

The problematic place seems to be the output stage that feeds the power amp, through shielded signal cables.
I had built the filter with TL072/TL074 on experiment board but the PCB was designed for two quads and I had only one TL074, so I used an LME49740 instead...

Think I burnt one of those sweet chips too :/
 
swapping a 3-4 MHz GBW op amp with a 50+ MHz GBW op amp is asking for trouble - likely the pcb and decoupling aren't designed for the faster part - especially trying to use the pDIP package at low/unity gains you might see in fliter/crossover circuits

and the higher speed does increase sensitivity to C load, often requiring the added isolating series R - or ferrite bead

also it is often not so good to replace a fet input circuit with a bjt op amp - often the noise will be higher because the impedance of the parts in a circuit designed for a fet input op amp are too high to take advantage of the lower Vn spec of the bjt
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.