EEVblog #922 - Mailbag

[EEVblog]

More Mysterious Mailbag


SPOILERS:
Tektronix / Promax RFM90 SignalMini Receive Signal Level Meter teardown
A mysterious FM receiver that transmits!
RoutaBoard prototype board matrix http://routaboard.com/
IntegriFuse USB Protection
https://www.indiegogo.com/projects/not-availible-usb#/
Digilent Analog Part Kit
http://store.digilentinc.com/analog-parts-kit-by-analog-devices-companion-parts-kit-for-the-analog-discovery/
Digilent PmodIA Impedance converter
http://store.digilentinc.com/pmodia-impedance-analyzer/
Raspberry Pi Proto Boards:
http://www.briandorey.com
https://www.youtube.com/user/briandorey
DC-AC mechanical vibrator inverter
RED Digital camera battery charger teardown


Datasheets:
https://cdn-reichelt.de/documents/datenblatt/A300/ST62T25CB6.pdf
http://pdf.datasheetcatalog.com/datasheet/toshiba/8.pdf
http://www.analog.com/media/en/technical-documentation/data-sheets/AD5933.pdf

[Towger]

High Speed Vibrator:

[vk3yedotcom]

On the oscillating FM radio, the first test I'd do is adjust the tuning and see if the 95 MHz changes.  If so it could indeed be the local oscillator - especially if it moves over a 20 MHz span as the tuning is adjusted between 88 and 108 (should also be 10.7 MHz removed from the dial frequency).

If it stays the same as the tuning is varied then that lets the local osc off the hook.  Maybe it's another stage that's spuriously oscillating. Waving your hand inside might give some clues. RF preamps sometimes self-oscillate but I'm not sure if cheap consumer grade receivers always have these. 

[djQUAN]

Anyone noticed when Dave tore down the RF signal level meter...

I was looking at the wires that looks like it leads to a small speaker and noticed those are solid wires and are plugged in the PCB and Dave was moving the PCB all over.

Thinking this would not end well and wanted to shout "unplug it!" and yep, both eventually broke. 


In the charger board, the black wires leading to the transformer appears to be a thermistor glued onto it for over temp sensing. Seen it several times on car inverters.

[wojteks]

Regarding the RED Charger:

As a camera assistant, I work with these chargers almost every day. It is true, they are very intelligent chargers. Due to the properties of Li-Ion cells, it takes 2-3 times quicker to charge a battery pack 0-80% (Constant Current mode) than it does to charge it 80-100% (Constant Voltage mode). The RED charger, in order to give you the minimum turn-around time for batteries, will charge Batt 1 to 80%, then Batt 2 to 80%, then will switch back to Batt 1 for CV mode, charge it from 80 to 100%, and then do the same for Batt 2. In addition, the industrial Lemo connector at the front can be used to power your camera directly.

Other trivia on this system:

The RED charger uses a standard Sony V-mount/V-lock mounting plate for the batteries, the same as the camera. This is pretty much the industry standard, you can power pro video monitors, LED panels, cameras and other gimmicks, all through a v-mount battery.

The two outer pins are for main power + and -, two of the three middle pins are defined as BAT_CLK and BAT_DAT for an I2C-like communication between the battery and charger/camera, and the remaining pin I believe is for temperature monitoring. The battery packs contain a PCB with intelligent charge monitoring, equalization and a physical thermal fuse.

The "RED Brick" Li-Ion batteries are rated at 153Wh, although compatible batteries can go up to 230Wh (15.6Ah) for a Beillen BL-BP230. The RED Epic camera can draw up to 10 amps, according to the specs, and (from my own experience) will shutdown around 12V.

[TheSteve]

There are techniques to speed up the 80-100% charge that involve pulsing high current. I know of one charger which uses this technique to achieve fast charging all the way to 100% full.

[Tek_TDS220]

Regarding the clock-radio, there is a lot of potential for oscillation in a receiver front end: big amplification and many opportunities for feedback. 

Please do a segment on the AD5933!  It's an amazing chip and fairly easy to use, especially with the Digilent demo board.  I've hooked mine up to a headless Raspberry Pi and I'm storing data from frequency sweeps.  However, I'm having trouble interpreting the data. 

[Dingo]

The AD5933 sounds great except for stopping at 100 ohms - when the target impedance for most interfaces is 50 ohms or occasionally 75 ohms...

[aromring]

The interesting fact about the featured FM clock radio is that when tuned to an FM station it works fine. The audio plays OK with relatively little noise. The distinct buzz is heard when tuned in between stations.

Anyway, there is an interesting story leading to me sending the radio to Dave. I bought it about 12 years ago and gave it to my son who put it in his room. He liked being awaken for school by either his favorite FM station, or CD music, as opposed to an annoying alarm clock. About 6 years later our remote garage door openers started acting strangely. It was really difficult to open the garage from a distance longer than about 6-7 meters. Back then, I had no idea whatsoever about electronics, radios, etc. You can guess my profession from my avatar. Hence, my first step was to change batteries in all remotes which did not help. Then check/restart all receivers, and so on. To make the long story short, after few days of no progress I was pulling my hair out, on the phone with manufacturer's support, trying this-and-that, to no avail. My wife was joking: "What evil spirits possessed our garage!" 

Finally, in one of the support calls a nice lady mentioned they had a few cases of radio interference. At first I dismissed it: "There are no transmitters nearby", but decided to test the idea anyway. I started turning off individual circuit breakers and - bingo - the moment I turned off the power to my son's room the garage remote started working OK! Therefore, it was a matter of turning off individual appliances there. I think the radio happened to be the first one to test. Thus, the joke of the day was: "Oh, it's the radio that is haunted"

For some reason I decided not to throw it away. Perhaps, I had an idea of playing a prank on our friends by offering it as a "gift"? Ha ha. Last year I picked electronics as my new hobby and - you guessed it - I started watching the excellent EEVblog. Last month I pulled the radio from my junk drawer and, instead of disassembling it for parts, tested it with SDR noticing the shown patterns. I thought it might be an interesting curiosity to all EEVblog viewers, and you know the rest of the story.

I only hope Dave is curious enough to find out which part has failed.

[Falkra]

Lovely coils inside the Tektronix RF strengh signal.   
Loved the haunted clock/radio/alarm. 

Very good mailbag episode. Cool and crazy stuff. 

[Tek_TDS220]

Aromring:  It must be that there is some feedback causing oscillation.  One common way for this to happen is in the power supply that feeds the various stages.  Often they have an RC network at each stage for isolation, and if the small electrolytics go bad, oscillation can occur.

Dingo:  The AD5933 is essentially a vector network analyzer on a chip, although Analog Devices doesn't advertise it that way.  You're right that the specs for impedance range and frequency are limited.   Still, it's a fascinating part.

[trophosphere]

The AD5933 sounds great except for stopping at 100 ohms - when the target impedance for most interfaces is 50 ohms or occasionally 75 ohms...

The impedance range can be readily modified by adding an external amplifier to the input of the chip. Take a look at the application notes for the AD5933.

How to Configure the AD5933/AD5934
Measuring a Loudspeaker Impedance Profile Using the AD5933

[cdonges]

Can you please put the mechanical inverter on an oscilloscope? I think it would be awesome to see the waveform.

[Red Squirrel]

I second the inverter on the scope!

That, and I want to hear that radio, should do a recording of that.

[aromring]

Tek_TDS220: You are onto something because the signal coming from that "haunted radio" is modulated at 120 Hz - I mention it in my letter and for a brief moment you can see the audio spectrum. 120 = 2*60 Hz, where 60 Hz is the mains frequency here in the States.

Red Squirrel: Your wish is granted: I am attaching the signal recording in .wav format, audio track, and audio spectrum.

[twice11]

Aromring:  It must be that there is some feedback causing oscillation.  One common way for this to happen is in the power supply that feeds the various stages.  Often they have an RC network at each stage for isolation, and if the small electrolytics go bad, oscillation can occur.

I also would suggest to hunt for haunted (or simply broken) bypass caps. They might be dried electrolytics, but typical radial electrolytics are no good at 100MHz anyway, so bypassing would better be done by ceramics. If they "forgot" to back the bypass cap with a ceramic cap, it might be the cause for oscillation, but you should consider a mechanically broken ceramic cap, too.