This post will be a bit of a rant about robust design principles and making a product that won't fail under real world use. But it isn't directed at anyone in particular -- agedhorse's post just put me into a state of musing about the elements of robust circuit design.
edit: As I proofread this post, I realize that it has actually turned into a rant about Mesa, so I guess it is directed at someone. To be fair, it would also apply to any other circuit designer who makes the same kind of design mistakes that Mesa made.
Just a heads up that this may not be a good idea for amps that use the current steered through the LEDs as part of the switching logic in a footswitched amp. I know our amps are not the only ones that do this, and changing the current (brightness of LEDs) can affect the operation of the switching logic as it's based on a voltage shifting constant current loop. This is a very effective, robust and troublefree method of discrete logic management.
I understand the rationale for the use of switching logic. But what I've never understood is why engineers will fail to protect the switching logic from damage by putting components of the logic circuit on the outside of the amp where they are at-risk for being subjected to trauma.
Bass amps are subject to wear and tear and contact on the road. It's a fact of life. If you go mounting switching logic components on the front panel of an amp, that could be a recipe for disaster -- One good hit on a front panel LED that has the switching logic passing through it, and BAM! your switching logic is busted. Now the amp's features can't be switched.
The smart manufacturers aren't putting the logic LEDs right on the face panel, because the face panel constitutes an exposed, threatened environment. It makes more sense to keep the logic LEDs mounted inside of the protective confines of the chassis, and to use them as backlighting, rather than as exposed elements.
For those LEDs that are mounted directly on the front panel, or any other externalized part of the chassis, it's just dumb to route internal control logic through a component that's mounted where it's accessible to the outside world, because external contact / damage to that component will interrupt the logic circuit.
A smarter idea would be to confine as many as possible of the switching logic components to the interior of the chassis, and to make the externalized components
controlled by the switching logic, rather than a
part of the switching logic. That way if you bash-in an LED, you only lose the indicator lamp that's controlled by the switching logic, rather than losing the switching logic itself.
Robust design parameters require that when an indicator lamp fails, that only the lamp fails, and that a logic circuit isn't compromised by a lamp failure. Granted, we're not talking about military-grade reliability, but I honestly think that some of the designers out there only think about the task of designing their circuit, and don't pay a lot of attention to making the actual product bulletproof in the field.
I'm one of those guys who thinks that switching logic passing through a front-panel LED would be a bad idea. It creates a failure mode that doesn't have to exist.
Here's an example where MESA really shot themselves in the foot:
I ran across this exact sort of problem with a MESA amp that has LED channel-activity indicators on the footswitch. The genius that designed the switching logic routed the logic signal for every channel through it's own LED that's on a footswitch. For Pete's sake, the footswitch is deigned to be stepped on, and the LED is mounted right there. Murphy's Law tells us that if you put switching logic in a place where it's going to get regular contact with THE BOOT, then the switching logic is going to fail.
Indeed, that's what happened. The frail little LED in it's plastic mount couldn't take the abuse of an enthusiastic thrashing guitarist stomping on it, and the poor little LED broke-off one of it's hair-thin leads, causing the logic circuit to fail. The result is that the amp was stuck in one channel. The switching logic would no longer switch channels, and the guitarist was stuck with a 3 channel amp that was stuck in lead mode.
A smarter design would put the footswitch LED under the control of the logic circuit without actually being a part of the logic control circuit. That way, when the LED gets thrashed under the boot, the indicator lamp fails to work but the switching logic in the amp keeps on working, and the user can finish the gig. (It's better to lose an LED lamp than to lose a whole amp.)
If anything, this sort of thing teaches us a lesson in intelligent design. It's a bad idea to put logic circuitry in an interface where it should be expected to fail. I think that in the case I'm describing, the design engineer made a critical error by failing to think things through. Maybe he was a smart EE who wasn't a musician, and he just didn't understand why it would be a bad idea to run the amp's control logic through a footswitch LED. In cases like this, it pays to have people who actually use this type of gear on the design team, or to at least beta-test the design with real musicians.
