It needs to be correctly loaded - 600 ohms and those rails, with 50% voltage out (33% Class AB power). I'd be quite surprised if fakes survive, most genuine parts have problems in these tests (and end up in the Chinese supply line). With no load you can't expect to be actually stress testing the part (only breakdown voltage). Operating means a (any) typical application, a 1KHz sinewave is a reasonable test signal.
The NE5532P (from TI datahseet version SLOS75H, circa 2004) is specified to 150C with Rth (j-a) of 85C/W. With ambient of 40C, it should easily survive 1.29W of output with no added heatsink. You can work out what load and output level you need to hit to get close to this number. A 600 ohm load (on each output) will result in 666mW of dissipation each channel. Two will be slightly more than our 1.29W, so you can increase the load a bit (680 ohm resistors is about 1.17W, a close enough number with some safety built in), or back off the output level a bit.
Your circuit needs to be very stable for this to work properly, you'll be at the ragged edge of the chip's ability. Sockets for example are absolutely not a good idea, and your high end PSU will come in handy.
The NE5532P (from TI datahseet version SLOS75H, circa 2004) is specified to 150C with Rth (j-a) of 85C/W. With ambient of 40C, it should easily survive 1.29W of output with no added heatsink. You can work out what load and output level you need to hit to get close to this number. A 600 ohm load (on each output) will result in 666mW of dissipation each channel. Two will be slightly more than our 1.29W, so you can increase the load a bit (680 ohm resistors is about 1.17W, a close enough number with some safety built in), or back off the output level a bit.
Your circuit needs to be very stable for this to work properly, you'll be at the ragged edge of the chip's ability. Sockets for example are absolutely not a good idea, and your high end PSU will come in handy.
Its driving two cathodes of grounded grids stage.
I thought the part was ‘too hot’ failed the 3 sec touch but it was still chugging along, low thd and all.
powered down, cool and re-test, no problems.
Not much overshoot either on 50khz square wave.
The thd is with tube ‘in the loop’. So i’m happy with it.
-
I thought the part was ‘too hot’ failed the 3 sec touch but it was still chugging along, low thd and all.
powered down, cool and re-test, no problems.
Not much overshoot either on 50khz square wave.
The thd is with tube ‘in the loop’. So i’m happy with it.
-
Attachments
Well i’ve smoked/damaged a few grade A conforming parts from mouser-digikey, two in same lot. Can’t dare go beyond spec. Kinda disappointing, maybe the new parts cutting corners in fab.
I’m looking at standardizing on the OPA445bM fet input for hybrids and NE5532 for cheapie intermediary stages. No 5532 have failed yet. Seems ruggid, even the $0.04 units.
Worldwide the opa445 looks very available. FFt looks super on both.
Adiós Amigos
I’m looking at standardizing on the OPA445bM fet input for hybrids and NE5532 for cheapie intermediary stages. No 5532 have failed yet. Seems ruggid, even the $0.04 units.
Worldwide the opa445 looks very available. FFt looks super on both.
Adiós Amigos
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.