Reading driver displacement with radar!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've read about using Doppler radar to read loudspeaker displacement before. There was an article in Audio Amateur once. So I tried myself using a Mattel Hot Wheels Radar Gun (J2358) and dismantled it for the waveguide and tried it myself.

It works nice. This is 30Hz being pushed too far into excursion-based distortion on one of my folded horns with an Eminence HL10c
An externally hosted image should be here but it was not working when we last tested it.


Please excuse the sawdust. This is the waveguide in my fabbed holder with a weighted base. The radar is passing through 2 layers of plywood.
An externally hosted image should be here but it was not working when we last tested it.


Limiter control circuit coming today. I'll add a thermal monitor as well for a complete protection system.
An externally hosted image should be here but it was not working when we last tested it.


See you all at Burning Amp this Sunday!
 
Last edited:
Hey dude. LTNS. The only trick appears to be a little inverse square compensation needed. As the driver moves closer to the waveguide, the output level of the mixer signal increases causing a visible 2nd H distortion. A little magic with a 4 quadrant multiplier and the receive signal looks right again. The pre-amp does need about 70dB of gain though.

What's interesting to me is that while the wavelength of 10.525 GHz is only 28.2mm, the displacement I'm measuring is just under that. So even though a frequency to voltage conversion is normally used for determining velocity with the doppler technique (like a car travelling @ 55MPH with police radar), I just get a phase output as I'm not crossing a wavelength boundary. And the phase output is directly displacement without the need of any fancy conversions.

Placement of the waveguide is critical.
 
Last edited:
Afraid that's a bit above the electronics kung fu I currently possess. My skills pretty much end with knowing which end of the soldering iron to hold.... Though I understand the basic concept of doppler radar and the issue at hand (measuring a back and forth motion), I sincerely doubt I could make it work. Very cool work. I want one.:D

Been so crazy busy with work and other projects all summer I've not had much time for speakers. I've not done anything cool with subs since the sub for my brother in March. As usual - more ideas than money or time.
 
I know the feeling! I'm only doing this project now cause I got laid-off. This idea for a realistics-based limiter for subs has been floating in my head for at least six months.

The idea is to push EQ for depth, not taking into account excursion. Once the driver limits are known for excursion, I can setup my ol' six-pack and let nature be my limiter. Maybe I'll be lucky enough to setup properly corner loaded one day, then 1/2 space the next. So EQ for good depth below my corner at reduced volume, and let the limiter manage what the coupling I get from the space.

I haven't put the thermal circuit together yet, but being able to monitor coil temperature in real time is just the sweetest thing ever. Imagine a readout at the FOH desk showing you power compression!
 
Probably cheaper at a local toy store that has some than from Amazon.com. Or there's garage door sensors, too, that use doppler. Or there's even building your own waveguide from an old coffee tin or whatever. The osc and mixer circuit, I have no clue on at the moment, but probably isn't that difficult to make except for the tuning tools needed, which probably don't come cheap.
 
Last edited:
a dumb thought floating through my head, what does the backemf from a sub look like once coil goes out of the gap, and past xmax? In the servocontrol from rhythmic, they use a separate sensing coil to control the servo... just wondering without that sensing coil if you can accomplish a similar feedback loop for your brown note without the sonar?

although, I have to admit, this is way geekier than making a 2.4ghz antenna out of a pringles can !!

btw, ever given any thought to v-coupling your AT's? (3 and 3, facing toward the point of the V. All you'd have to build is the part for them to fire into.... It'd get you closer to 30hz Fc)
 
Last edited:
Hi Jim. I'm not searching for the brown note, but I am sick. I had a headache all this afternoon from blasting 30Hz in my garage. I still have it. So I took the time to put in a mouser order and work-out on paper the control circuit for the gain reducer. I think I saw the same WiFi antenna article, too, with the pringles can.

I don't know if there is a way to calculate backemf from a coil that is being driven. An undriven coil acts as a pickup from the motion through the cap. If the drive voltage becomes rather different than the pickup coil...

Velodyne used a piezo element in their cones as an accelerometer to determine displacement. They were notorious for breaking. There's also laser interferometry but that's too exact and has a resolution down to 1/4 the wavelength of the laser which for red is 161nm. That's crazy small. Or there's bouncing a laser beam off of a reflecting surface like tin foil to an X detector. The six-pack already does 34Hz F3, 29Hz F10 in 1/2 space.

12kHz in sound has the same wavelength of 10.525 GHz of RF. Diagnostic ultrasound is around 2 MHz. So that'll need the more complex FtoV conversion.

I'm not looking for any other method, this seems to work as is without any fuss.

If I wanted to V-couple, I'd make some T-36 cabs to drop the HL10s into. Am I trying to go lower, no. What I'm trying to do is to use real world limits as limiter inputs. Not some inference from an excursion/power/frequency chart to set your assumed safe high-pass. The six-pack already does 34Hz F3, 29Hz F10 in 1/2 space.
 
Last edited:
I think I saw the same WiFi antenna article, too, with the pringles can..
Well, we took that to a ridiculous extreme a few years ago. Seems a local telephone company was trenching rural fiber, and providing internet access/telephone. About 10 miles away was the closest it would get to my dad's farm. So, we set a wap11 on top of a windmill with a yagi, pointing line of site to a 100' silo. On top of that silo was yagi's with 3 wap11's back to back. One to the local school 18 miles away, one to my dad's farm close and one to my brother's place about 10 miles away. The receiving antenna at my brothers place is the one I think you'll be interested in... Seems there was an unused birdview aluminum C-band dish... and not wanting to spend money on yet another yagi... well a makeshift 2.4ghz waveguide on that dish worked incredibly well... Not sure the fcc would have been terribly happy with any of that --- but it worked well for about 2 years, till a better internet solution could be had. (often wondered what the gain on a 10' aluminum is....)

Your sonar effort reminds me of that. Using simple tools to do something cool. great idea, thanks for sharing.
 
schematic ready

Rough schematic of what I showed at BAF2009. Sorry for the delay for those who asked.

Replace R15 with a pot to adjust threshold. U3 is used as a compensator to adjust-out a 2nd harmonic distortion caused by inverse-square. As the driver comes closer to the antenna, the signal goes up, thus causing a "fat bottom" to the signal. U3 compensates by applying an adjustable square function. It probably isn't need, but my eyes told me I had to have it.

Active differential input and active servo floating output missing.. apply as required. Or transformers if you like ;)
 

Attachments

  • WooferLimiter.pdf
    56.4 KB · Views: 137
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.