opamp rolling the correct way?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I may have been a bit harsh.

It's most definitely rewarding and perhaps even a fundamental skill in electronics understanding power distribution in complex audio systems. Break it down to the smallest unit of processing (a single opamp) and give it the best environment possible to do its thing. You will start viewing opamp datasheets in utilitarian way. Having done the homework I find it somewhat inexcusable to see so many designs skip on this area due to parts count or just applying whatever the first examples of the datasheet does. Although being somewhat absolute on this topic may have something to do with myself having spent so many years in failure mode. In other words listening failing opamps doing all things distortion, oscillation and placebo. This isn't an advanced guru mediation, but comes up undergraduate level in most universities I believe. And way before any serious radio frequency work.

I know there's an actual job out there that blends accounting with EE and JohnRoberts has certainly shared on his experiences on this. I would be curious on why cheap and bad design still gets applied on todays SMD scale. Doesn't seem it would make much difference per single PCB to apply the brute force and perfect approach in quantities above 10000.
 
I could probably talk about this for several six-packs worth.  8) Actually being part accountant is what engineering really is about. It's not about just making a circuit do XYZ, but doing it without leaving a lot of money on the table, that the competition can use to beat you on price.

I fear examples of bad implementation will probably get even worse in the future as more and more real (analog) design is accomplished inside the ICs (those guys are doing the real deal), so more practical real world design work is connecting the dots between cookie cutter, cut and past blocks from app notes.

If you look at app notes for switching PS controller ICs you will often see recommended PCB layouts too. Apparently so many customers screwed up the layouts that they felt compelled to give them a cut and paste template or the layout too.

I used to manage an engineering group and there is always pressure from above to not reinvent the wheel, but use proved circuits, over and over. Of course any engineer worth his salt, lives to improve even his own previous designs, let alone some app note poop.

The most egregious example of design by accountant was the old Crown CE1000/2000 power amps. They literally took the schematic from an old DC300 (even older Crown amp), lost the DC coupled part, and then repackaged it using modern SMD parts and plastic power devices.  The PCB layout/repackage engineer may have been a SMD guy but he wasn't a power amp guy and the mistakes were spectacular. For one example the power devices usually use several watt emitter resistors, even though in use thay never see that much power. The fireworks occur when the power transistor shorts as they inevitably do, and now instead of just replacing the power device there was a charred hole in the PCB where the SMD emitter resistor was.  :eek: :eek:

It isn't rocket science but it is science.

JR
 
Kingston said:
This isn't an advanced guru mediation, but comes up undergraduate level in most universities I believe.
Alas, many pseudo gurus are unaware of these basic principles.  Today, the important buzzword is to use only stuff Hand Built by Virgins from Solid Unobtainium.  Cheapo Electrolytics must be evil and left out completely .. regardless of what the datasheet says.

At Calrec, we frequently argued over grounding issues though we had been making mixing desks at the highest level for many years.

I still agonise over grounding and PCB layouts for even simple preamps and pore over anything that purports to say anything new about grounding.  Alas, there is a lot of liquid BS running around.

But this Millienium, I learnt something new about evil ground planes.  You STILL have to ensure your currents run where you want them on ground planes.  If you can't route them as though there wasn't a ground plane, you will have problems.

JohnRoberts said:
I used to manage an engineering group and there is always pressure from above to not reinvent the wheel, but use proved circuits, over and over. Of course any engineer worth his salt, lives to improve even his own previous designs, let alone some app note poop.

I know there's an actual job out there that blends accounting with EE and JohnRoberts has certainly shared on his experiences on this. I would be curious on why cheap and bad design still gets applied on todays SMD scale. Doesn't seem it would make much difference per single PCB to apply the brute force and perfect approach in quantities above 10000.
An engineer is someone who can do for 2 bob what any fool can do for a quid - Anon

Non est tantum facile - moi

I was R&D Manager at Wharfedale for nearly half my life and spent much time asking young engineers what they expected to gain from their supa dupa new complexity.  I'm old school in that I count bits and begrudge even 2 extra resistors if I don't see any advantage.  But of course, R&D Depts. are infested with ..

"technocratium"...the element that has no protons and all free electrons - it is uncontrollable (even in the presence of administratium) and has the ability to keep on designing things even after all requirements have been met ....
 
maybe ill use these parts to try and design a preamp instead of popping and pushing opamps into my mixer.  sounds like ill fail often and learn more designing from scratch. better use for the chips perhaps.
 
Back
Top