Microprocessor-monitored tube amp at SparkFun

[cantgetnosleep]

Thought this was kinda cool. The guys at SparkFun set up a tube amp with a microprocessor to monitor the voltages and shut down in case of over voltages.

http://www.sparkfun.com/commerce/tutorial_info.php?tutorials_id=89

 

[Svart]

I guess it's kinda cool although we've been doing things like that in the commercial world for years.

 

[PRR]

> a tube amp with a microprocessor to monitor the voltages and shut down in case of over voltages.

Wimp.

The high voltage is the fun part. You only live when you stare death in the face. And adding silly-state parts "to protect" good hard metal-vacuum devices is just lower reliability: build it right and it probably won't die, nor cost a lot to repair.

I'm at least as scared around "low"-voltage HIGH-current systems. Even an old Dyna 120 will burn your ring into your finger. Modern kiloWatt amps have lethal voltages at copper-splattering amperage.

 

[shabtek]

If you didn't:start--listening to PRR

You only live when you stare death in the face.

this is the mentality one must have to manually jumper around the fused switch of a 333kVA xfmr in order to replace it and avoid interruption of 'the juice'.
'tis a cheep trill here in da windy city

edit: oh yeah I thought that amp looked like a bunch of work for little reward...good exercise though. would be better utilized if there were rare/more tubes/ costlier failure potential; common power tubes and current production OPTs
perspective= >thousands of amps built with similar audio circuit, many still intact work, pretty reliable without the bells and gristle

 

[PRR]

> manually jumper around the fused switch of a 333kVA xfmr in order to replace it and avoid interruption of 'the juice'.

This happened in my school a few weeks back:

"...staff was doing maintenance on a fan unit in (University main server-room).  As a result of this work, an 800 amp circuit breaker was tripped.  In an effort to reset the breaker additional actions were taken..., and as a result of these actions, one of the busbars was damaged."

I'm glad I was across campus when that 800A busbar got "damaged".

My estimate of the peak available power: lash all 33 engines at the Indy 500 together, and try to stop them with a screwdriver. Damage happens.

"We were told that all power to the building was going to be cut within 15 minutes...  Graceful shutdowns were not much of an option, and we literally had to begin pulling plugs. ... By 10:30, all non-critical equipment had been turned off to help keep the room from overheating and damaging other more critical components."

"The effort included fabricating components by an out of town parts manufacturer."

Upshot was NO email for over 15 hours. Also most official University websites were dead.

There is of course considerable discussion about how this happened. My observation: while Google has shiny new cargo-pack server systems, our server-room dates from the early 1970s (yes, before Al Gore invented the InterNet, even before ARPAnet), and a lot of the infrastructure must be "quaint".

 

[Andy Peters]

Thought this was kinda cool. The guys at SparkFun set up a tube amp with a microprocessor to monitor the voltages and shut down in case of over voltages.

http://www.sparkfun.com/commerce/tutorial_info.php?tutorials_id=89

That's called "housekeeping."

a-