Test Equipment Anonymous (TEA) group therapy thread

[0culus]

There's no such thing as a digital circuit...only analog circuits forced to have 2 stable states.  Sooner or later most digital people end up learning that the hard way.

[mnementh]

Yup... how many times I've had to explain to people (especially since I got into multirotor RC aircraft with digital FCs, ESCs and numerous digital accessories that make a modern RC quadcopter into a RF HOSTILE environment) that EVERY DIGITAL SIGNAL MUST FIRST EXIST IN THE ANALOG DOMAIN.

95% of the time all I get back is that look like a tree full of owls. 

mnem
But it's that 5% who DO grok that will eventually carry the torch... so we must carry on. 

[bd139]

Distance, time, energy, entropy are all discretised so it’s digital under that analogue 

Disclaimer: may be more turtles further down.

[tggzzz]

There's no such thing as a digital circuit...only analog circuits forced to have 2 stable states.  Sooner or later most digital people end up learning that the hard way.

I repeatedly make a similar point, emphasised by the caveat "... unless you are doing photon counting or femtoamp circuits".

I also note that the receiver interprets the analogue voltage (or current) as a digital signal.

[0culus]

Distance, time, energy, entropy are all discretised so it’s digital under that analogue 

Disclaimer: may be more turtles further down.

 

See, at the scales I care about that doesn't affect me. I leave that for the physics nerds. 

[tggzzz]

Distance, time, energy, entropy are all discretised so it’s digital under that analogue 

People can understand that current is quantised, once they think, remember physics lessons, and realise those lessons actually have relevance in their world.

[worsthorse]

I am still digging through the box I stuff I saved from the dumpster. Last night I found a Trimble Thunderbolt and this:



A couple of hours of interwebs digging revealed no useful information. Figure if anyone knows what this is, that person is spending time on this thread. Here are some more photos:







Any ideas?

[bd139]

Distance, time, energy, entropy are all discretised so it’s digital under that analogue 

People can understand that current is quantised, once they think, remember physics lessons, and realise those lessons actually have relevance in their world.

Yes exactly. When you look at things like TLC NAND FLASH solid state disks it’s or extreme relevance for example. All they do is squirt and store a discrete electron count in a floating gate.

It really does help with understanding fast and small things if you think of electron shuffling.

[bitseeker]

(Or, if all that seems too much research, just get some old HP catalogs and start working your way through them, getting one of everything.)

I was going to say that catalogs are even more dangerous than manuals, but refrained from doing so because just mentioning it is a slippery slope. Too late now! *Swish*

So what is this "one of everything" you speak of? You can't have just one. It never works out like that.

[PA0PBZ]

Any ideas?

http://www.jancorver.org/info/aprs/files/lassenskii_man.pdf

[URI]

There's no such thing as a digital circuit...only analog circuits forced to have 2 stable states.  Sooner or later most digital people end up learning that the hard way.

That's true for electronic circuits using semiconductors.

It's not true for electromechanical switches aka relays.       

[kj7e]

Nice teardown!  What is the serial number prefix of the supply?  (first four digits)  I'm trying to date the caps, and it looks like 8434 for the big ones (34th week of 84), and maybe 8447 for the 1500uF one - can't quite make it out.  Though that seems old given the excellent condition of the supply - the s/n prefix would establish the earliest it could have been manufactured.  (And I like how you hide your vodka in plain sight on the bench in that innocent looking squeeze bottle... )

-Pat

Serial number starts with 2451, so its post 1984 but that only reflects the version, due to the rocker power switch I believe this one was manufactured after 91'. 

The good news, I bought a pair of NOS Sprague 1500uF 40v snap-in's on eBay that arrived today and they test out perfect!  Testing at 1700-1800uF each @ 1KHz and 1560uF + or - a few at 120Hz.  Just got done ramping up the voltage on them slowly, showing very little leakage so I will be putting these in for sure.

[kj7e]

Sorry to hear about all the bad caps though... that BOM is gonna be pretty pricey. 

mnem

Not too bad really, picked up the NOS 1500uF 40v Snap-in on eBay for $13 total.  I have 4 brand new 12000uF 50v United Chemi-Con main filter caps on order from Mouser, $13 each.  The other small caps are all Vishay/BC from Digikey and came to about $15 total.  Well worth it if you ask me.

[bd139]

That’s not bad actually. One thing that puts me off the larger end Hp supplies is having to replace the coke can sized capacitors in them

[PrecisionAnalytic]

Stepper motors from printers are A) not powerful enough (except to move a laser) to do anything useful and 2) always use some idiotic mounting system that does not play well with anything except molded plastic bosses and stamped sheet metal and Ω) almost globally use oddball windings that aren't compatible with any stepper driver that will play nicely with common ardu-based 3DP/CNC controllers.

"Build it rigid, flat and square, THEN add motors & electronics".

Just consider that $250-350 as your tuition; cost of entry. It really is the cheapest way to get into "learning mode" quickly, rather than spending days and weeks in "pulling your hair out" mode. The thing with all of this is that it's all "hobbyist" and "experimental" gear; it's not like a Craftsman table saw that comes with instructions and even a "User Guide" so you have some idea where to start. Most everything requires trawling the web and usergroups to get knowledge, then build your own instruction set that works for YOU.

Please... learn from my mistakes.  You will save yourself oodles... just oodles... of wasted time, money and aggravation. 

mnem


And if you know all this, just consider my post as a PSA for the next guy who doesn't. 

I wasn't thinking last night for some reason about what I saw when going to the freezer in the garage afterwards... all the free treadmills I picked up and tore down for material to use to make the larger 4'x4' CNC system.  Yeah... I figured... why not use treadmill motors, metal and components since each usually has 2 motors, one for the treadmill and one for the incline.  They come with controllers also.  The motors and controllers work great for mini-lathe upgrades too.

Here's the inspirational video that started me down that path of gym equipment re-purposing for humanity:

[worsthorse]

Any ideas?

http://www.jancorver.org/info/aprs/files/lassenskii_man.pdf

That's what I was looking for. Mahalo!

[tggzzz]

There's no such thing as a digital circuit...only analog circuits forced to have 2 stable states.  Sooner or later most digital people end up learning that the hard way.

That's true for electronic circuits using semiconductors.

It's not true for electromechanical switches aka relays.       

It is when the mechanics become stiff, or the contacts damaged,or the coil is damaged, or wiring is damaged.

Or, of course, if you look at the voltages with a scope, particularly during switching.

[PrecisionAnalytic]

"Build it rigid, flat and square, THEN add motors & electronics".

mnem


And if you know all this, just consider my post as a PSA for the next guy who doesn't. 

Reminds me since I've worked on Coordinate Measuring Systems implementations, validation, method development, etc... I had a vision also of if I find a granite slab/table that I could also make this into a Laser, Touch/Feather Probe and most likely Vision coordinate measuring system too.  Damn... I was slippin into darkness.  Nuts is I just saw a system that can make an awesome bed for the lathe/mill upgrade and the other parts will work as the larger at least 4'x4' CNC system with granite table.  I'll try to get pictures next time I'm there.

[xrunner]

It's time -3D printing has arrived (almost) to the lab!

[Cubdriver]

(Or, if all that seems too much research, just get some old HP catalogs and start working your way through them, getting one of everything.)

I was going to say that catalogs are even more dangerous than manuals, but refrained from doing so because just mentioning it is a slippery slope. Too late now! *Swish*

So what is this "one of everything"you speak of. That never works out like that.

Good point - you want two, in case one breaks.   

-Pat

Edit to add - you don't know slippery slope until you get nearly ALL of the catalogs.  At that point, forget it - you're done for...

[mnementh]

Stepper motors from printers are A) not powerful enough (except to move a laser) to do anything useful and 2) always use some idiotic mounting system that does not play well with anything except molded plastic bosses and stamped sheet metal and Ω) almost globally use oddball windings that aren't compatible with any stepper driver that will play nicely with common ardu-based 3DP/CNC controllers.

"Build it rigid, flat and square, THEN add motors & electronics".

Just consider that $250-350 as your tuition; cost of entry. It really is the cheapest way to get into "learning mode" quickly, rather than spending days and weeks in "pulling your hair out" mode. The thing with all of this is that it's all "hobbyist" and "experimental" gear; it's not like a Craftsman table saw that comes with instructions and even a "User Guide" so you have some idea where to start. Most everything requires trawling the web and usergroups to get knowledge, then build your own instruction set that works for YOU.

Please... learn from my mistakes.  You will save yourself oodles... just oodles... of wasted time, money and aggravation. 

mnem


And if you know all this, just consider my post as a PSA for the next guy who doesn't. 

I wasn't thinking last night for some reason about what I saw when going to the freezer in the garage afterwards... all the free treadmills I picked up and tore down for material to use to make the larger 4'x4' CNC system.  Yeah... I figured... why not use treadmill motors, metal and components since each usually has 2 motors, one for the treadmill and one for the incline.  They come with controllers also.  The motors and controllers work great for mini-lathe upgrades too.

Here's the inspirational video that started me down that path of gym equipment re-purposing for humanity:

Ummm.... because you need stepper motors. The only thing you can use the treadmill motor for is the router spindle, which you'll spend more making a collet adapter than a whole bolt-on spindle assembly will cost. Unless you already own a machine shop with mill and lathe, which why the eff are we even talking about this? 

And looking at that guy's whole series.... there's literally MONTHS of building there; I say this as someone with considerable experience doing welding for a living, and building stuff JUST LIKE THAT table he made as solenoid-operated fabrication jigs.

Furthermore, yeah he says he spent $300 building that... I suspect he's not counting a lot of stuff. I see easily $150 ($60 claimed is total BS unless he bought it all salvage) in just raw steel, because you don't get to count only what you use, you count ALL the steel you buy for the project.

Then of course add the thousands of dollars of shop he's using to make that $300 CNC table... and you're not looking so cheap at all.

So yeah... start with a kit. Save yourself LOTS of misery. Once you actually know some of what you're doing, THAT is the time to go balls & braces making other stuff work.

"Build it rigid, flat and square, THEN add motors & electronics".

mnem


And if you know all this, just consider my post as a PSA for the next guy who doesn't. 

Reminds me since I've worked on Coordinate Measuring Systems implementations, validation, method development, etc... I had a vision also of if I find a granite slab/table that I could also make this into a Laser, Touch/Feather Probe and most likely Vision coordinate measuring system too.  Damn... I was slippin into darkness.  Nuts is I just saw a system that can make an awesome bed for the lathe/mill upgrade and the other parts will work as the larger at least 4'x4' CNC system with granite table.  I'll try to get pictures next time I'm there.

Yeah, for 95% of CNC fab work, that kind of precision just is not necessary. Accurate to hundredths of an inch is plenty; level to within a few mm will get you close enough to be inside the leveling adjustment of any modern system.

8020 build system is effing MAGICAL. I made assembly jigs and conveyors and even jig KITS to be shipped across the country to sites where literally maintenance workers would be assembling it. The best part is that due to its nature, as long as you cut it with the prescribed cutting equipment (the $249 chopsaws from Lowes don't even come close to accurate enough) and jigs, the parts literally square and center themselves up as you draw the bolts down. 

mnem
*couch-tuber time widda family: Mortal Engines*

[beanflying]

It's time -3D printing has arrived (almost) to the lab!

Great first choice 

Treat the Magnetic bed a bit carefully the surface is fairly soft so set your bed height carefully (err on the high side). I have damaged mine using 0.1mm layers in one spot by dragging the nozzle. Also the Capricorn tube and better fittings is really worthwhile https://www.captubes.com/

A few upgrades I like on mine. Print them in this order while you get it and your slicer settings sorted. In order of Easier to Harder.

Keep your cables nice - because it feels good 

Display cover because exposed electronics are dumb

Bits Tray, great for tools bits etc

Beanflying Ender Teflon lined Filament guide

Improved part cooling duct using stock fans

[mnementh]

   It's time -3D printing has arrived (almost) to the lab!

YESSS! 

It has a dragon right on it... it HAS TO kick ass! };=)~~~<

Actually, I toyed with the idea of the Ender myself, even when I was shopping the Tornado. I need the 340mm x 400mm capacity for a few RC projects that are still in the cogitatin' stage. that, and the Tornado's 120V bed are what sold me... though I'm finding it is a mixed blessing with the mirror glass tiles.

Again... 8020/4020 build system FTW!!! 

Oh, yeah... DON'T put the gawddamn filament holder on top of your frame like they show it in the adverts. 

That just amplifies any ghosting issues you may have.

mnem
*Filamental*

[factory]

(snip)
Here are some pictures of the insides instead, I do have an untested rectifier bulb to replace the later resistor diode MOD bodge underneath (the serial number is in the range of the last ones to have had a hollow-state rectifier);

That Dale power resistor has seen A LOT of current and looks mighty crispy.  I would check that circuit for a partial short or some other reason for the excessive current.

You've spotted part of the MOD bodge then, I think the later 738BR has the resistor (R73) bolted to the chassis for a start. It will get checked next time I work on it, it's listed as 100Ω 20W and drops the voltage same as the original tube rectifier would have done (clearly they didn't want to change the transformer).

David

[xrunner]

Great first choice 

Treat the Magnetic bed a bit carefully the surface is fairly soft so set your bed height carefully (err on the high side). I have damaged mine using 0.1mm layers in one spot by dragging the nozzle. Also the Capricorn tube and better fittings is really worthwhile https://www.captubes.com/

YESSS! 

It has a dragon right on it... it HAS TO kick ass! };=)~~~<

Thanks guys. I went to a friend of a friend's house and he had two of these and two Prusas going! He is making a full size C3PO exoskelaton for a person to fit into and needs parallel print processing.

I've looked at many build videos and will be building it tomorrow. Also noticed that it has some factory improvements since some of the build videos - such as the control electronics enclosure has the fan on the bottom now. All else seems to be of high quality. Should be a fun road ahead!