Don't forget you promised me pictures of the ugly display surgery.
No TEA here this weekend. The lady is here, currently sleeping and I'm sneaking some time on the computer. She's still sore from the fight incident on Wednesday but recovering nicely. Tomorrow (Saturday) will be a busy day taking her shopping.
Hooray Ms. med! Hope you were a good "old man" and snuggled her to sleep.
I can't believe that. You're building electronics gadgets by the dozen and repair stuff frequently, yet you do not have a 120 Ω resistor? Seriously, I never thought anybody posting here wouldn't have at least a complete E6 series of resistors from 10 Ω to 1 MΩ. But then, 240 Ω saves a wee bit of current.
I have a full Royal Ohm dev kit. I should say that in this case the issue is not that I don’t have them but more it looks like I accidentally put the 120 ohm ones in the bench sweepings box and I couldn’t be bothered to go and find them
240 because you can do the maths easier in your head. 1.25 volts over 240 is about 5mA. So I need to drop 3-1.25=1.75 volts. 1.75 / 5m is 350 (5 goes into 17.5 three and a half times) so pick a 330. Close enough! Build, measure, 3.0004v. Close enough. Sorted. The minimum stable standing current of an LM317 is about 10mA but that’s only if you have a capacitive load attached so 5mA was fine here. 120/150 is better if you have that.
Thus the whole thing was at the lazy engineering maxima. Also because calculator and phone were buried under parts and schematics.
I have at least a dozen "baggie kits" of axial lead and SMD resistors I've never ever bothered to sort into the RAACO boxes like I should have; they all reside in a plastic shoebox along with several baggie kits of trim pots, 10-turn pots and carded Radiddio Shank resistors rubber-banded together for so long I've had to change the bands at least a dozen times.
At the bottom of the shoebox is my "fuckit" pile... a big old swarm of bench sweepings, random single resistors that have fallen off of the sprag or card, all sorts of random scavenged power and precision resistors and pots, etc.
I've gotten in the habit of FORCING myself to at least LOOK there for a few minutes for any
single resistor I need before rummaging in the baggie kits; but only over the last few years.
We won't talk about the pitiful state of my capacitor collection... I just did a massive haul to the dumpster of capacitors I KNOW were
at least 10-20 years old.
Ashamed I didn't do it 10 years ago.
But I still haven't braved the hallowed halls of Mouser, DigiKey and Newark to find replacements.
@bd139, you are correct not a single DC blocking cap in sight, such is the design these days, it doesn't even have any speaker protection circuit either unlike the Kenwood receiver I have that checks the speaker connections before switching the output relay on to connect the amp to the speakers. Its all about penny pinching these days.
Not sure what part it is handling the output there (assume it's some kind of either push-pull arrangement or class D switcher) but the assumption that the output is never going to be DC because nothing is ever going to fail is a dangerous one for the manufacturer. Big thumbs down on that design
I don't think it's even penny pinching here but probably just verbatim copying of the output IC manufacturer's datasheet which probably doesn't mention anything like that because datasheets only carefully mention the "within spec" side of things and inside the SOA it won't blow anything up. Similar to the old original rev 1 LM386 datasheets which fail to mentoin the zobel network on the output which caused many a little shitty amp to go kaboom from oscillation way above audio frequencies.
"National does not assume any responsibility for use of any circuitry described"
This is why when designing things engineers should look at the output and input circuits inside the black box ICs so they can understand what the IC designers forgot or left out intentionally because of cost or practicality reasons. In fact I think it should be a requirement that you're not allowed to use an IC until you've build at least an approximation of the circuit out of transistors at least once in your life. Side issue; same with programmers. They should know roughly how the system they are using is constructed from the bottom up.
Step 1: MIT 6.002 "the abstraction". Worth a watch if you haven't seen it before. I wish my lecturers at the time were this good and informative:
Anyway well off tangent there.
I tend to agree as well... more a matter of lazy engineering (in the bad way) by way of using the bare minimum shown by the manufacturer and possibly a little of the good old "What? We can shave a few cents per unit here AND increase turnover by 10% over the next 5 years? WIN-WIN!"
Stef Murky brand of management decision-making process.
I like this professor... he reminds me of very few great professors I've been lucky enough to have; that not only teach you the rote material, but teach you to THINK in new and larger ways. To grok in fullness. I had one such professor in my junior year at high school; when he taught math he both made it interesting and enlightening, AND he helped you understand how the fundamental concepts build upon each other.
Any professor who can make SELF-DISCIPLINE literally part of the equation qualifies as a great professor.I had another such professor at SACC; without his incisive insight into the abstractions of modern networking, I'd never have grokked the OSI model. *Sigh*
@mnementh
Here are the 2 pages that cover the circuit of interest, amp and the speaker/headphone circuits and as you will see the speaker selection switch most definitely sits within the path between power amp and the speakers.
Oh, I believe you... especially after you traced it back to the IC. It's funny how "general design ethic" seems to go in cycles... something new comes along (Like multi-channel monolithic amplifier ICs you can mute individual channels on, making soft-switching cheap & easy) and for a while, everybody is using it... then maybe the bottom falls out of some component like a certain type of switch, and suddenly the old-school way is notably cheaper again so regains favor.
There could be (and should be) a whole SEPARATE school of actuarial and statistical science devoted to just this aspect of modern manufacturing... it would help the few smart leaders we have left position our workforces to be producing what is really needed rather than bouncing around from one manufacturing fad to the next like lemming ping-pong balls.
mnem