Power supply and quiescent current

[Rolox]

A question from a noob (me):

Consider a 2 way active crossover with just 5 double op-amps (originally NE5532)

What do I need to look at to make sure the power supply could feed op-amps with higher current demands without over-heating? (I'd like to try discretes).

Voltage regs are LM317 - 337 in TO220 package. No heatsinks.

 

[rayma]

You should get at least 100mA without heat sinks, and perhaps your supply could give even more current
with individual heat sinks bolted on the regulators. That depends on the unregulated input voltages and the
capacitors used (and maybe the transformer).

 

[Mark Tillotson]

NE5532's can't drive less than about 500 ohm loads (without distortion rising), so at +/-15V you'd never expect more than 25mA per opamp stage (plus quiescent), and in higher impedance situations that will be correspondingly less. For 10 stages the worst case is 0.3A per rail, and probably a lot less.

 

[Rolox]

You should get at least 100mA without heat sinks, and perhaps your supply could give even more current
with individual heat sinks bolted on the regulators. That depends on the unregulated input voltages and the
capacitors used (and maybe the transformer).

I would like to use discrete op-amps; lacking a complete datasheet I can only have a wild guess at what a dual opamp package consumes but I'd consider 60mA quiescent current per package to be a safe assumption; multiplied by 5 that's 300mA quiescent, not even mentioning dynamic current...

Is there anything you can tell me about the feasibility of the project, looking at the attached pictures?

Is there anything bad that could happen if, say, I just plug the 5 discrete duals and see what happens? I suppose the voltage regs will just shut down if they overheat, but will there be damage occurring to something in the circuit or to the PSU?

I'm sorry if my questions are stupid - this is meant as an experiment (comparing a professional crossover with IC op-amps, from a big name, with a no-name but well built cheap crossover from China with equally Chinese discrete op-amps. But I don't wanna blow up anything.

 

[Rolox]

Well I can't upload any picture for some reason :-/

 

[Rolox]

Ahhh here are the pictures

 

[mchambin]

Voltage regs are LM317 - 337 in TO220 package. No heatsinks

I would add heatsinks.

 

[rayma]

I would add heatsinks.

May not be enough room for sinks. Also the transformer winding may not be large enough for the extra loading.
You could try it, but check the voltage across the regulators and make sure it is at least 3VDC, and preferably more.

 

[Mark Tillotson]

Is there anything bad that could happen if, say, I just plug the 5 discrete duals and see what happens?

You mean apart from to your bank balance? Seems an illogical approach given the lack of data on most of these products and the fantastic performance levels achievable with much cheaper parts. That money is probably better spent on improving room acoustics and loudspeakers.

 

[mchambin]

Once powered, check how hot the regulators get, transformer too.

 

[Rolox]

You mean apart from to your bank balance? Seems an illogical approach given the lack of data on most of these products and the fantastic performance levels achievable with much cheaper parts. That money is probably better spent on improving room acoustics and loudspeakers.

That's another debate I don't wish to get into - I've been using IC opamps countless times I just want to experiment discretes for once, plus they were cheap.

 

[Mark Tillotson]

Ah, these chinese ones I guess - it would be good if you can document their performance relative to standard opamps - power consumption, noise (voltage and current and flicker), distortion at 2k and 600 ohms.

 

[Rolox]

I couldn't gather much.
The only available specs (at least on the sellers page, maybe it would be possible to contact the factory but, who is the factory? haha) are:

Ultra low distortion: 0.0008%
Low noise: 4nv√Hz
BJT-Input: 30pA
High speed: conversion rate -120V/us bandwidth -8MHz
High open loop gain: 120dB (600Ω)
Power supply range: ±12V~±18V (standard ±18V)

and the fact that they are unity gain stable and protected against phase issues.

They do actually offer a quite extended range of discrete op-amps with gradually increased bandwidth and open loop gain, and proportionally increased price. I got the cheapest - risk is minimal and I thought it might be the easiest to use in a NE5532 circuit. They advertise this as being the ideal NE5532 replacement, wich probably means nothing anyways.