Official Gramps' Memorial Eternal OT Thread

#65,402
#65,402
NorthDallas40 said:
Nah @AM64 and @VolStrom get it too.

But with the stupid đŸ’© done to this one thr same design done digitally would be untapped too.

And on thinking on it today I want to say I did a simple quick desktop review about a year and a half ago of this design (they’ve been screwing around a long time on this thing) before they released it for layout and told them to remove basically the same damn list of parts đŸ€·â€â™‚ïž

Oh well. More money for me.
Click to expand...
You're making me think that the younger generations of U.S. engineers aren't qualified to do too much more than adjust the settings on an iPhone.
 
  • Like
Reactions: AM64, NorthDallas40 and VolStrom
#65,403
#65,403
Vol423 said:
You're making me think that the younger generations of U.S. engineers aren't qualified to do too much more than adjust the settings on an iPhone.
Click to expand...
I recall saying something similar to that one day to a manager. I believe I said something like "If they can't look it up on their phone, they don't have a damn clue what to do".
 
  • Like
Reactions: AM64, NorthDallas40 and Vol423
#65,404
#65,404
Vol423 said:
You're making me think that the younger generations of U.S. engineers aren't qualified to do too much more than adjust the settings on an iPhone.
Click to expand...
So where I work(ed) the way I describe it is the mean skill level seems pretty much the same but the standard deviation has gotten way out of whack. Some of the kids I have worked with in recent years are motivated, driven, well prepared, and highly intelligent. But at the other end there are an equal amount of useless dead weight pieces of **** I won’t waste the time on.

I’ve made it clear to them that mentoring is a two way street. The protege isn’t the only one with a say. The teacher has to think they’re worth the effort or it doesn’t work. And I have always picked my own protĂ©gĂ©s
 
  • Like
Reactions: tntar heel, AM64 and VolStrom
#65,405
#65,405
NorthDallas40 said:
Some of the kids I have worked with in recent years are motivated, driven, well prepared, and highly intelligent. But at the other end there are an equal amount of useless dead weight pieces of **** I won’t waste the time on.
Click to expand...
I know what you've dealt with. We had a summer intern come through one year and he was a POS. Everything someone tried to show him, he just blew it off like he already knew it and couldn't be bothered. He spent 95% of his time glued to his phone. He was a minority, and apparently he thought he was a golden child and could do no wrong. He ended up getting fired halfway through the summer and never knew what a huge opportunity he gave up.
 
  • Like
Reactions: AM64 and NorthDallas40
#65,407
#65,407
NorthDallas40 said:
So they had me working on a simple old analog motor current amplifier my first week. They’ve apparently been struggling with getting it functional for about a month. From looking at the test data I said it looked unstable. Took some time to make a simple model that validated my assumption. In an earlier review of the schematic I’d already identified some likely evil parts. I updated my model with removal of those parts and voila the model says its stable! Told the engineer which five capacitors to remove and test it Monday I’ll check emails Monday afternoon. Had the damn laptop not burped I’d likely have finished this in less than a full week after they’ve dicked around with it for a month. I think I’m earning my keep 😎
Click to expand...

Unless something is obviously smoked and the power supply is working, just go to capacitors and redo any funky looking solder joints ... not that you can find either in most stuff these days.

I had a hearing aid go bad last week, and thought I was finally going to have to break down and get new ones. Really bad static (like an AM radio off station). I took it apart (nothing repairable of course) but cleaned up the guts and blew some canned air over it. Then dried out the desiccant in the microwave and put them back in the box with dry desiccant overnight. Working like a charm again. Now I'm trying to get old SW going to see it I can still program them. Old SW and obsolete HW, but it seems to talk to Windows 10 despite what others say. Sometimes I think electronics just need to know you care enough to fix what ails them.
 
  • Like
Reactions: NorthDallas40
#65,409
#65,409
Anyone here had artificial turf installed in their yard? I have a neighbor a few houses down that had their backyard covered in it, and it looks damn good. My dog has absolutely killed the grass in my backyard and I'm thinking about putting it in my yard too.
 
  • Like
Reactions: tntar heel
#65,411
#65,411
AM64 said:
Unless something is obviously smoked and the power supply is working, just go to capacitors and redo any funky looking solder joints ... not that you can find either in most stuff these days.

I had a hearing aid go bad last week, and thought I was finally going to have to break down and get new ones. Really bad static (like an AM radio off station). I took it apart (nothing repairable of course) but cleaned up the guts and blew some canned air over it. Then dried out the desiccant in the microwave and put them back in the box with dry desiccant overnight. Working like a charm again. Now I'm trying to get old SW going to see it I can still program them. Old SW and obsolete HW, but it seems to talk to Windows 10 despite what others say. Sometimes I think electronics just need to know you care enough to fix what ails them.
Click to expand...
No it was a poor design. Or a or poorly considered higher level design I should say. The designer sprinkled caps on each stage to limit the amplifier bandwidth arbitrarily. Fine for a local stage but when you chain all those stages together the resulting phase loss is oppressive. The amplifier will naturally roll off on its own it’s an integrating loop. My “fix” was as simple as pulling the safety blanket caps off and let the amps run up to their gain bandwidth product. Boom, the resulting model prediction looked just like I told the guy to do it (over a year ago if it’s the design I think it is). We’ll see next week if that was all of it or if there are more rotten layers under the onion.
 
Last edited:
  • Like
Reactions: Matt2496 and AM64
#65,412
#65,412
VolStrom said:
Anyone here had artificial turf installed in their yard? I have a neighbor a few houses down that had their backyard covered in it, and it looks damn good. My dog has absolutely killed the grass in my backyard and I'm thinking about putting it in my yard too.
Click to expand...

No but I wouldn't be opposed to it, some of that turfgrass looks pretty real from a distance and no more weedeating would be nice.
 
  • Like
Reactions: VolStrom, Carl Pickens and AM64
#65,416
#65,416
And yes all you cretins. As least right now I’ll admit I’m enjoying using the lump of đŸ’© between my ears again. Damn I was rusty on solving some basic amplifier equations tho
 
  • Like
Reactions: Vol423 and AM64
#65,417
#65,417
NorthDallas40 said:
So they had me working on a simple old analog motor current amplifier my first week. They’ve apparently been struggling with getting it functional for about a month. From looking at the test data I said it looked unstable. Took some time to make a simple model that validated my assumption. In an earlier review of the schematic I’d already identified some likely evil parts. I updated my model with removal of those parts and voila the model says its stable! Told the engineer which five capacitors to remove and test it Monday I’ll check emails Monday afternoon. Had the damn laptop not burped I’d likely have finished this in less than a full week after they’ve dicked around with it for a month. I think I’m earning my keep 😎
Click to expand...

We were called in to look at a bearing pedestal problems - between a steam turbine and the pump it drove. The plant had been operating for years with a few problems in this system. Because it was a safety system (high pressure coolant injection) in a nuclear plant, it was a safety system that had to be tested periodically - far more test runs than actual use - fairly normal in nuclear plants. Anyway after a series of nonconclusive tests we broadened out and decided to take a look at even the turbine controller, water in the steam lines, etc because it just wasn't a vibration problem.

Turns out the system was designed to start with both the start valve and governor valves wide open - basically like starting up from every stop with the foot on the brake, engine screaming with the throttle on the floor, and popping the clutch. The really weird part was after the start, the governor valve went closed and ramped open in at a sane pace. We got a lot of flak about messing with a safety system when we told them what had to be done - like open the start valve with the governor valve closed and then ramp it open reasonably if you don't want to rip out bearings, crack the pedestal, or slam less than decent quality steam through a cold turbine. How some things pass the design stage is just beyond comprehension.
 
  • Like
Reactions: NorthDallas40
#65,421
#65,421
NorthDallas40 said:
No it was a poor design. Or a or poorly considered higher level design I should say. The designer sprinkled caps on each stage to limit the amplifier bandwidth arbitrarily. Fine for a local stage but when you chain all those stages together the resulting phase loss is oppressive. The amplifier will naturally roll off on its own it’s an integrating loop. My “fix” was as simple as pulling the safety blanket caps off and let the amps run up to their gain bandwidth product. Boom, the resulting model prediction looked just like I told the guy to do it (over a year ago if it’s the design I think it is). We’ll see next week if that was all of it or if there are more rotten layers under the onion.
Click to expand...

Do they ever do any testing like to measure transfer function (gain and phase) on systems like the one you are talking about? Actual testing rather than calculated result? I'd say electronic design is probably a lot more reliable than mechanical design, but you'd be amazed at the differences we often found between how things as built responded compared to design calculations. We used modal analysis a lot to determine actual vibration modes in components vs calculated values; I don't know if there's comparable testing and analysis for electronics, but it would perhaps reveal weird feedback loops and instabilities.
 
#65,422
#65,422
AM64 said:
Do they ever do any testing like to measure transfer function (gain and phase) on systems like the one you are talking about? Actual testing rather than calculated result? I'd say electronic design is probably a lot more reliable than mechanical design, but you'd be amazed at the differences we often found between how things as built responded compared to design calculations. We used modal analysis a lot to determine actual vibration modes in components vs calculated values; I don't know if there's comparable testing and analysis for electronics, but it would perhaps reveal weird feedback loops and instabilities.
Click to expand...
I will be doing just that next week. Swept sine source so I get really good coherence but it takes a bit more time. But 
 it needs to be stable before you can do that.

This poor design never had as chance as built unfortunately. Hopefully I have corrected that and I can turn the knob up to 11 on it next week.
 
  • Like
Reactions: AM64
#65,423
#65,423
Rookies. My Marshall tuber goes to 11

xitbrexbmu270zhadmt1.jpg
 
  • Like
Reactions: NorthDallas40 and AM64
#65,425
#65,425
NorthDallas40 said:
No it was a poor design. Or a or poorly considered higher level design I should say. The designer sprinkled caps on each stage to limit the amplifier bandwidth arbitrarily. Fine for a local stage but when you chain all those stages together the resulting phase loss is oppressive. The amplifier will naturally roll off on its own it’s an integrating loop. My “fix” was as simple as pulling the safety blanket caps off and let the amps run up to their gain bandwidth product. Boom, the resulting model prediction looked just like I told the guy to do it (over a year ago if it’s the design I think it is). We’ll see next week if that was all of it or if there are more rotten layers under the onion.
Click to expand...
This brings back memories of amplifier biasing and bandwidth calculations and modeling feedback amplifier topologies in SPICE. Those were fun times.
 
Last edited:
  • Like
Reactions: NorthDallas40 and AM64
New posts
You must log in or register to reply here.
Advertisement





Back
Top