Electrical Engineering - 2017-11-20

« first day (2580 days earlier) ← previous day next day → last day (2359 days later) »

12:11 AM

@PlasmaHH Electric Can Opener

3 hours later…

3:19 AM

hey there @uhoh

user image

2

Q: Why is TESS' high gain antenna made of undulating BLACK fabric rather than metal?

note: 1 after I placed the bounty, I think I may have figured it out, but not absolutely sure. If you read through carefully, you might too! (there are clues in the newest image I found in Facebook, below and in the EXCELLENT video linked below that) note 2: I will not place another pro-active b...

radio-communication antenna tess

2

Has +200 bounty expiring in 3 days...

unfortunately, can't help you with it but good thinking to drop it here, hopefully someone sees it :)

1 hour later…

4:52 AM

@Shalvenay mouser.com/Search/…

I think I found what you were talking about. That in combination with that VCO I linked, an MCU, an amplifier and an antenna should do it, as far as I can tell. Is that IC what you were talking about?

@Allenph one problem with that part -- Cypress is going to quit making it soon

Saw that. I found a couple alternatives...but this is indeed what you were suggesting I find, right?

@Allenph that kind of IC, yeah

Now the trick is figuring out how to debug the circuit with a 50MHz scope...

@Allenph do you have some sort of measurement receiver or spectrum analyzer at hand?

5:02 AM

I have a scope, a multimeter, and a crappy Chinese function generator.

The scope is an old analog Tektronix.

ah

I'd keep your eyes peeled for a spectrum analyzer then

you're in hardcore RF territory right now

where your test points are coax connectors

Indeed. Writing your

Indeed. Writing your own spread spectrum protocol is...hard.

every square mm of real estate is a potential component

and 50ohms is love, 50ohms is life

@Allenph so yeah, head back to your favorite source of used test equipment and keep your eyes out for a spectrum analyzer

5:18 AM

Would that even be helpful considering with spread spectrum my location will be in pseudo-random frequencies? Not to mention the confusion of everyone's routers and other assorted devices.

I have no idea what your resistance comment is about.

@Allenph the former can be handled by setting the sweep on your spectrum analyzer wide enough to see what's going on with the SS, the latter is easy though: coax from TX out to specan-in via an attenuator

Ahhh. I see. The cheap handheld ones show images with an antenna.

But now I see the antenna is attached by coax.

Such expensive. Wow.

re: 50ohms -- that's the characteristic impedance of the coax cable and connectors used in radio and test work

(75ohm coax is reserved for video and TV signals)

I see. At least the amplifiers are cheap. Going to run like 20 bucks a set of ICs.

and your PCB traces at such high frequencies basically need to be flat coax (aka stripline or microstrip)

so they wind up being 50ohms too

5:34 AM

Wait...so all this work I spent making etching acid and getting stuff to make PCBs is useless?

5:56 AM

After googling a bit...I don't see any problems except a little less power.

3 hours later…

8:46 AM

@uhoh I am pretty sure I saw a similar question somewhere a while ago...

1 hour later…

10:08 AM

@PlasmaHH I left a similar notice about this question here five months ago. This is a new bounty.

...but it's the same question.

that might explain it...

do you know if the undulation is in reflective material, or if it is actually transparent?

3 hours later…

12:42 PM

@Allenph you can still etch microstrip into double sided board yourself

its more of a layout technique than anything else

1 hour later…

2:06 PM

@Allenph It honestly depends on whether you like the process of etching your own PCB, or if you can take advantage of the shorter turn time.

I'm much happier sending my boards out. You get vias, soldermask, and a guaranteed space/trace for a very reasonable price.

If you're doing RF, you might also be able to get some panels of material at lower cost than a custom board using them

2 hours later…

4:34 PM

@W5VO I collect hobbies. Chemistry is one of them. It gave me a side project to make some cupric chloride...which I also wasted money and time on.

@Shalvenay From reading about it, it just seems like I could lose some output power...and not very much at that. Is that not the case?

The guy I get my boards from does double sided boards even cheaper than single-sided. Why? I don't know. Regardless, I just bought 50 square feet of single-sided board from him so that sucks.

5:20 PM

Also...where do you guys hunt for electrical test equipment? Those spectrum analyzers are too rich for my blood with Christmas coming up.

5:32 PM

If you don't need really high performance, you could probably get one of those RTL-SDR packs to work

otherwise ebay and word of mouth

.... I might know someone who's got a whole bunch of CMU200's cleaned and tested by mid December....

@Asmyldof There's no way I could afford such a machine unfortunately. Thanks though.

@W5VO I'll look into that...my uncle actually just suggested looking at the same.

I mean, what kind of budget are we even talking about?

Used ones run about $500-600 on ebay for name brand

5:51 PM

No where close to that. Maybe $200 max.

Looks like I can't even afford a HackRF One or bladeRF.

6:02 PM

a RTL SDR will probably give you useful qualitative measurements

but not that useful for matching or quantitative power measurements

It's not range or power I'm concerned with. As long as it goes 10 feet it's more than it will ever need to be. It's analyzing the circuit and eventually the output of my spread spectrum code that's the problem.

6:53 PM

If I'm reading 13MΩ between to points on my circuit (with a consumer grade multimeter), does that effectively mean "no connection"?

Depends on how you're expecting things to be connected. High impedance inputs may exceed 100MΩ, while dirt might push an unconnected node to less than 1MΩ

I think it's effectively "not connected" in my context. Still baffled as to why my 74HCT245 is outputting 4.8V unloaded, rather than its 5.0V input voltage.

4.0V, that is.

I'm beginning to wonder if the chip itself is hosed.

7:11 PM

Anyone wanna buy 6 cleaned CMU200's yet?

#askingForAFriend

I just worked out my problem 🤣

In the immortal words of Blackboard from Mr Squiggle: "Upside down! Upside down!"

@RobertAtkins . . and translated means ?

I had the damn chip plugged in to the board upside down.

Probably not very good for it, I suspect

Oh goody, and now the Arduino software can't see my RFduino on USB :-(

7:29 PM

@Asmyldof If you have a price maybe I can sell my liver.

Maybe.

8:08 PM

@Allenph Human liver, even at rush rates should be good for three or four of them

He's going to make someone buy all 6?

Half a liver might be enough for the same if you're willingly going to submit to the guy with the blood test kit that rings the bell tomorrow

That reference is lost on me.

Easier to find a match if .... my acquaintance of an acquaintance of a neighbour of someone I once met has some blood-work a week or so in advance...

1 hour later…

9:38 PM

@RobertAtkins All bets are off when you're dealing with chip inputs

or trying to go through them

9:54 PM

I tried another one and it was fine, so I guess I fried it. Oh well...

10:14 PM

Arrrr...statistics

@ThePhoton I remember you. You tried to answer all my terrible questions during my last foray into EE. I believe a series of bad questions about OpAmps.

10:43 PM

@ThePhoton Based on a sample size of 2? Welcome to the club....

@Asmyldof Take 100 current readings. Calculate mean and st.dev. Do it again...totally different st.dev.

Well, not "totally different", but jumping around by a factor of 4x.

@ThePhoton Build better set-ups

@Asmyldof But just when I needed this set-up, somebody else in the office happened to be building dozens of them to send off to the factory.

...

It was all so convenient

« first day (2580 days earlier) ← previous day next day → last day (2359 days later) »

Transcript for

Nov '1720

Electrical Engineering

A place to talk with friends from the EE community about vacuu...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile