Test Equipment Anonymous (TEA) group therapy thread

[med6753]

In other news. It's currently 41 degrees F (5 degrees C) here. Won't be long before the first frost. Guess I should pull the A/C out of the window and stash it in the closet until next year. Just last week I had to use it one afternoon when the temp soared to 80 F with high humidity.  The landlord hasn't yet turned on the furnace to provide heat. He isn't required to by law until October 15th which is tomorrow. So far it's still comfortable in here and I do have a space heater just in case.

Very jealous.  67 degrees on the way to 85 today.  As the saying goes, 'fall is in the air' said no one in Florida ever.

I could never live there. I don't do well in the heat/humidity and I like the change of seasons. Then I get sick of snow and can't wait until Spring. 

[Cerebus]

If it weren't for the influence of the Gulf Stream in the Atlantic,  Ireland...and you...would be in a deep freeze right now.

Like I said, I know the technical reasons, it's just the natural instincts that get a "Wha?" response. Rather like c being the speed limit, I know it, I can explain it, my head believes it, but somehow my gut doesn't want to believe it.

Boston: is ~8˚ or ~550 miles south of Penzance: where this counts as a lot of snow.

Edited to add: The point of picking Boston and Penzance is that both are ports and both are essentially on the Atlantic, even if it's called something different locally, so even the moderating effect of being at the seaside ought to be broadly similar (baring the Guld stream effect, which is what we all know really makes the difference here).

Oddly enough I am in a deep freeze now, I'm in the kitchen defrosting it.

[med6753]

Yep, I hear you. I was in Edinburgh in March for a week and I don't think it dropped below freezing once even at night. And although it wasn't warm during the day it certainly was a lot warmer than home which was down around 0 degrees F at night, barely 20 degrees F during the day, and 1.5 feet of snow on the ground.

[mnementh]

Keeping in line with random diversions on aviation, lol: https://www.aviation24.be/military-aircraft/belgian-air-component/air-force-f-16-destroyed-maintenance-collateral-damage-second/

Edit: and some more fail http://www.airforcemag.com/Features/Pages/2018/October%202018/Tyndall-F-22s-Left-Behind-Before-Michael-Hit-Possibly-Damaged-Beyond-Repair.aspx

BelgianWaffle sez "Watch where you point that thing; it might be loaded!!!" 

I remember similar arguments about the F-22 to those made nowadays about the F-35 over the idiocy of making a few obscenely expensive aircraft vs lots of "merely capable aircraft" and training pilots properly to fly them. I still feel it's a valid argument, but FFS, we need to find some balance in the muddle between the two.

We are making plans to keep B-52s in the air for the next 2-3 decades... this is a purely analog design that is already 65 years old, and the newest of which are 56 year old airframes. That's an aluminum airframe of stressed-skin design, which is literally dependent on the inherent flexibility of the material it's made of to keep the wings from snapping off in midair.

I'm only 50, and I'll tell you... I lost most of my inherent flexibility decades ago. 

At the same time, we have F-35 fighter craft that cost 25% more per copy (adjusted) than those B-52s did when they were new, and we can't keep them in the air for more than a few hours at a time. And we have to worry whether they'll BSOD and fly straight into a mountain with the pilot unable to do a thing about it.   

Like I said... SOME balance between the two... and that balance is not the Resident Chump solution of beleaguering the manufacturer until they shuffle the numbers around so you can SAY you made them "cut the price" to less than $100 Mil a copy. That's just pissing off your supplier, and they're going to make sure you pay twice as much in the end. 

Keeping in line with random diversions on aviation, lol: https://www.aviation24.be/military-aircraft/belgian-air-component/air-force-f-16-destroyed-maintenance-collateral-damage-second/

As I started reading the article at the link I was annoyed by the use of the of the weasel-word euphemism "collateral damage*" in the headline, until I read far enough to garner that it was a friendly fire** incident between aircraft inside a maintenance hangar.  Someone's not getting their Christmas bonus...

Edit: and some more fail http://www.airforcemag.com/Features/Pages/2018/October%202018/Tyndall-F-22s-Left-Behind-Before-Michael-Hit-Possibly-Damaged-Beyond-Repair.aspx

The disadvantage of having a military airfield so far from any realistic action zone is the belief that it's OK to build your hangers out of cardboard as they won't ever have to resist enemy action. (*cough*) Pearl Harbour (*cough*) like incidents aside, wouldn't it be wise to keep prevailing weather in mind, even if you aren't expecting enemy action? Ah, yes, "Military Intelligence".

*Military/police jargon for "we murdered or maimed innocent bystanders"
**Nothing friendly about it at all. If your 'friend' is shooting at you, they ain't yer friend no more.

"MAXIM 5: Close air support and friendly fire should be easier to tell apart."

Obversely: "There's a reason we don't give handguns to monkeys." *

In short, training FAIL.

If it weren't for the influence of the Gulf Stream in the Atlantic,  Ireland...and you...would be in a deep freeze right now.

That's okay... in a decade or two the Gulf Stream and the EAC will trade places, or worse, cease to exist at all.  Our kids & grandkids will get to experience the seasons in a way never seen since apes started walking upright. 

mnem
* Yet we give the nuclear button to an orange baboon with all his brain cells in his nutsack.

[Cerebus]

"MAXIM 5: Close air support and friendly fire should be easier to tell apart."

OK, I see where we're going with this. Let the shenanigans begin...



"MAXIM 11: Everything is air-droppable at least once."

[mnementh]

IIRC, the Maxim originally attached to that image was "MAXIM 34. If you’re leaving scorch-marks, you need a bigger gun."

Or maybe it was "MAXIM 25: If a manufacturer's warranty covers the damage you did, you didn't do enough."

It's been a while.

mnem

[Mr. Scram]

That F16 story is an instant classic.

[mnementh]

yup. One of these for sure.

mnem
*procrastinating-ily*

[bd139]

Seeing as there's nothing to buy, I made some more test gear. I like the DG1022Z I really do but it's expensive, noisy and a bit wanky sometimes. Thus looking around for something I can make, I found a design by W7ZOI for an "HP 8640 junior". This is a wideband oscillator, with divider (74hc74) and a bunch of filters. Don't have any variable capacitors and don't like them so I dug around and found a bag of 1SV149 varactors and a 10 turn pot, bodged them onto the front end of that, chucked it in an enclosure I had lying around and hey presto, an RF signal generator. Well half of one. No buffers, PA, dividers yet but it works pretty well sweeping 13-30MHz 

Did consider doing one with an Si5351A but I blew my last one up last week. Oops.

The original design is missing three things which I'm going to fix: ALC (will implement with "rf probe" sampler, opamp, TL431), readout (PIC+LCD - have both lying around) and sweep input (easy with VCO). Should kick out levelled 7dBm max when done. I'll add an external step attenuator for other powers required.

Ominous unlabelled case:



Gubbins. Not a lot to it:



Original article: https://www.cwtd.org/AnalyzeThis/8640jr.pdf

Edit: disclaimer. Will probably never finish it.

[GreyWoolfe]

I could never live there. I don't do well in the heat/humidity and I like the change of seasons. Then I get sick of snow and can't wait until Spring. 

I don't do well in heat/humidity either.  I am here because by ex-wife got me to move here.  She wasn't my ex when that happened.  I liked the idea of moving again even less than the heat.  Now, at this point, I am too old to start on the bottom rung of the ladder at another company and right now there is no other state program I can move to with my company.  If we ever actually get a chance to retire, it will be somewhere there is at least 3 seasons.

[Terry01]

Seeing as there's nothing to buy, I made some more test gear. I like the DG1022Z I really do but it's expensive, noisy and a bit wanky sometimes. Thus looking around for something I can make, I found a design by W7ZOI for an "HP 8640 junior". This is a wideband oscillator, with divider (74hc74) and a bunch of filters. Don't have any variable capacitors and don't like them so I dug around and found a bag of 1SV149 varactors and a 10 turn pot, bodged them onto the front end of that, chucked it in an enclosure I had lying around and hey presto, an RF signal generator. Well half of one. No buffers, PA, dividers yet but it works pretty well sweeping 13-30MHz 

Did consider doing one with an Si5351A but I blew my last one up last week. Oops.

The original design is missing three things which I'm going to fix: ALC (will implement with "rf probe" sampler, opamp, TL431), readout (PIC+LCD - have both lying around) and sweep input (easy with VCO). Should kick out levelled 7dBm max when done. I'll add an external step attenuator for other powers required.

Ominous unlabelled case:



Gubbins. Not a lot to it:



Original article: https://www.cwtd.org/AnalyzeThis/8640jr.pdf

Edit: disclaimer. Will probably never finish it.

Sorry for swearing but,,,,, HOLY FUCK!!

Did you just make that?
I need to swear again but i'll keep it in till I get off the post! I wish I knew how to make things like that. I probably never will know enough to make things like that but I can dream,,, right?

Nice build buddy! 

[mnementh]

(SNIP)
   Ominous unlabelled case:
Edit: disclaimer. Will probably never finish it.

Schlock Mercenary #000001

"Ommmmminous Hummmmmmmm...."

A perfect opportunity to recurse back to our earlier 70 MAXIMS discussion... 

mnem
"MAXIM 27: Don't be afraid to be the first to resort to violence."

[med6753]

I could never live there. I don't do well in the heat/humidity and I like the change of seasons. Then I get sick of snow and can't wait until Spring. 

I don't do well in heat/humidity either.  I am here because by ex-wife got me to move here.  She wasn't my ex when that happened.  I liked the idea of moving again even less than the heat.  Now, at this point, I am too old to start on the bottom rung of the ladder at another company and right now there is no other state program I can move to with my company.  If we ever actually get a chance to retire, it will be somewhere there is at least 3 seasons.

Isn't funny that an ex gets us to do things that we really, really don't want to do but we do anyway which is one of the many reasons why they are an ex. 

[bd139]

Did you just make that?
I need to swear again but i'll keep it in till I get off the post! I wish I knew how to make things like that. I probably never will know enough to make things like that but I can dream,,, right?

Nice build buddy! 

Thanks. I am only assembly worker here. The design is by Wes Hayward / W7ZOI. Angles strategically used not to show up the awful metalwork 

99% of what I make goes in the bin. Occasionally something works. I've got a now 5Kg box of blown up, fucked up and unfinished stuff

[med6753]

Seeing as there's nothing to buy, I made some more test gear. I like the DG1022Z I really do but it's expensive, noisy and a bit wanky sometimes. Thus looking around for something I can make, I found a design by W7ZOI for an "HP 8640 junior". This is a wideband oscillator, with divider (74hc74) and a bunch of filters. Don't have any variable capacitors and don't like them so I dug around and found a bag of 1SV149 varactors and a 10 turn pot, bodged them onto the front end of that, chucked it in an enclosure I had lying around and hey presto, an RF signal generator. Well half of one. No buffers, PA, dividers yet but it works pretty well sweeping 13-30MHz 

Did consider doing one with an Si5351A but I blew my last one up last week. Oops.

The original design is missing three things which I'm going to fix: ALC (will implement with "rf probe" sampler, opamp, TL431), readout (PIC+LCD - have both lying around) and sweep input (easy with VCO). Should kick out levelled 7dBm max when done. I'll add an external step attenuator for other powers required.



Original article: https://www.cwtd.org/AnalyzeThis/8640jr.pdf

Edit: disclaimer. Will probably never finish it.

Sorry for swearing but,,,,, HOLY FUCK!!

Did you just make that?
I need to swear again but i'll keep it in till I get off the post! I wish I knew how to make things like that. I probably never will know enough to make things like that but I can dream,,, right?

Nice build buddy! 

Terry, you are not alone. RF baffles me too.  And I am also impressed. 

Bd139....you better finish it! 

[bd139]

RF baffles me too.

solution (worked for me):

http://www.arrl.org/shop/Experimental-Methods-in-RF-Design-Classic-Reprint-Edition

+

https://www.elsevier.com/books/rf-circuit-design/bowick/978-0-7506-8518-4

+



 

[med6753]

Did you just make that?
I need to swear again but i'll keep it in till I get off the post! I wish I knew how to make things like that. I probably never will know enough to make things like that but I can dream,,, right?

Nice build buddy! 

Thanks. I am only assembly worker here. The design is by Wes Hayward / W7ZOI. Angles strategically used not to show up the awful metalwork 

99% of what I make goes in the bin. Occasionally something works. I've got a now 5Kg box of blown up, fucked up and unfinished stuff

But you at least have some grasp of RF and it's complexities. It drives me nuts. And yes, that BNC connector is crooked. 

 

[med6753]

RF baffles me too.

solution (worked for me):

http://www.arrl.org/shop/Experimental-Methods-in-RF-Design-Classic-Reprint-Edition

+

https://www.elsevier.com/books/rf-circuit-design/bowick/978-0-7506-8518-4

+



 

Ever hear......"can't teach an old dog.........."?   

[Specmaster]

Speaking as another person who is baffled by the complexity of RF, how do you make the connections on the copperboard that aren't ground, do you use some form of stick on pad or something because it clearly is not a printed circuit? 

[bd139]

Speaking as another person who is baffled by the complexity of RF, how do you make the connections on the copperboard that aren't ground, do use some form of stick on pad or something because it clearly is not a printed circuit? 

You start with a sheet of FR4 single sided substrate copper side up. Then you stick strips (cut with some Wiss shears) on to the surface with superglue. Solder parts on to the strips or the ground plane. Commonly called Manhattan construction.  You don't have to do it this way but it leads to a solid ground plane which is very important and doesn't require etching a board and looks relatively neat (which is important if you're OCD like me )

This isn't specifically for RF - even really slow circuits can do funny things without decent ground planes and construction etc.

Here's one I did earlier with a better angle. Looks pretty neat but in this case, performs terribly 



Edit: main reason is it's bloody cheap. 100x160 boards are about £1.40 from Rapid. Get through about 5-6 a year: https://www.rapidonline.com/rvfm-copper-clad-double-sided-fr4-fibre-glass-board-100-x-160mm-34-0820

[Wolfgang]

Speaking as another person who is baffled by the complexity of RF, how do you make the connections on the copperboard that aren't ground, do use some form of stick on pad or something because it clearly is not a printed circuit? 

You start with a sheet of FR4 single sided substrate copper side up. Then you stick strips (cut with some Wiss shears) on to the surface with superglue. Solder parts on to the strips or the ground plane. Commonly called Manhattan construction.  You don't have to do it this way but it leads to a solid ground plane which is very important and doesn't require etching a board and looks relatively neat (which is important if you're OCD like me )

This isn't specifically for RF - even really slow circuits can do funny things without decent ground planes and construction etc.

Here's one I did earlier with a better angle. Looks pretty neat but in this case, performs terribly 



Edit: main reason is it's bloody cheap. 100x160 boards are about £1.40 from Rapid. Get through about 5-6 a year: https://www.rapidonline.com/rvfm-copper-clad-double-sided-fr4-fibre-glass-board-100-x-160mm-34-0820

... you could use mini striplines with proper impedances up to 23cm:

https://electronicprojectsforfun.wordpress.com/prototyping/

scroll down a bit to see a tuned amp that works OK.

What counts is *electrical* instead of *neatness* beauty.

[bd139]

Indeed. However at <30MHz, feck it, neatness

I have actually just sent boards out with striplines on to JLCPCB as an experiment. Basically a 2.4GHz VSWR meter so I can play with wifi antennas. It's basically a directional coupler and schottky detector. No idea if it'll work but will be fun finding out

Incidentally all the design tools for this are shite I've found. Probably have to warez Genesys or something 

[mnementh]

Most of the China-direct sites are now offering small CNC router tables for cheap... I keep thinking all that's really needed for fast small-run PCB fab is some DECENT, reasonably accurate open-source software to "carve away everything that isn't a circuit" on DS FR4.

I've seen some startups TRYING to do it, but the actual solutions are no where near as "ready for prime time" as 3DP is, and they're crazy expensive for the half-baked product they do have.

I may actually live long enough to see something that works.

mnem
*p00t!*

[bd139]

I never saw the point of those CNC tables for PCBs. Too much complexity. I actually made a fair chunk of PCBs in my time, going back to secondary school, using photoresist + UV which is pretty cheap and easy. If it's a really simple one shot, just draw the damn thing out with a marker... works down to 0805 / SOIC at least....

Edit: this was a 3 hour turnaround including the design and construction...









Even the parts for that one were recycled off a duff HP scope, apart from the AD8307 which was aliexpress gold.

(forget controlled impedance etc - good enough for HF / low VHF mainly)

[Cerebus]

If you don't like gluing on the little connectivity islands to a Manhattan board, an alternative is to use a little spot cutter (like a tiny diamond concrete cutting core drill) to cut out the copper in rings to form the islands. Hard to find, easy to make if you've any workshop tool making experience. You lose the advantage of a complete ground plane, but that's often not as much of an disadvantage as some would have you believe. Indeed, it can be an advantage as each Manhattan island is no longer a little capacitor to ground.

An alternative to cutting out little squares for islands, is to use a heavy duty plier punch to punch little disks of PCB material, which you can then glue on