Has anyone tried ultrasonic distance measurement by PICAXE alone, not using SRF005 etc ?

Buzby

Senior Member
Hi All,

Couldn't do any work on this for the last two weeks, so today was my first 'distance capability' test.

Getting about 12m reliably outdoors, but the receiver needs to be pointing directly at the transmitter. I'm fairly certain that boosting the TX supply will get more range, but I'm not doing that until I'm more confident about the whole idea.

The beam spread of the transmitter seems to be about 90', so it looks like I'll need a circular array of receivers on the mobile, but again, I'm not doing that that just yet.

Next task is to get the distance measurement working.

That's more work in a hotel room !.

Cheers,

Buzby
 

Buzby

Senior Member
Well, I'm in my hotel room, and think I've discovered a problem.

The mobile unit is about 8m away, but the base unit is picking up reflections of the initial pulse from things in the room closer than 8m.

These reflections are as powerful as the response from the mobile, so I can't filter them out.

I now expect any room will have odd reflections, not just hotel rooms !. ( Even the floor could be closer than the mobile. )

However, the Marvelmind system works indoors, so they must have solved this issue.

The only way I can think it works is the way I posted very early on in this thread, the initial pulse is RF or IR.

If the mobile is triggered this way then there are no early echoes from the base unit, only late echoes from the mobile, and late echoes are no problem, as the direct signal will always be first to arrive at the base.

So it's back to the drawing board !.

I've not got any RF bits with me, so it's going to be another week with no results.

Cheers,

Buzby
 

hippy

Technical Support
Staff member
The mobile unit is about 8m away, but the base unit is picking up reflections of the initial pulse from things in the room closer than 8m.

These reflections are as powerful as the response from the mobile, so I can't filter them out.
This is why I was suggesting a delay between seeing a trigger and responding. If we exaggerate the delays, let's assume it's one second. The master sends out a pulse, any echo within that one second can be ignored; it is a reflection, not a reply -
Code:
           _
Trigger  _| |_____________________________________
                 _     _    _             _
Echo     _______| |___| |__| |___________| |______
            |                   |   |
          | `--- reflections ---'   `--- reply ---
          |     
          |------- 1 second ------->|
The actual delay doesn't have to be as long as one second, just long enough for all reflections to have occurred.

I am guessing the Marvelmind system is this -

https://www.unmannedtechshop.co.uk/product/marvelmind-robotics-indoor-navigation-positioning-system-starter-set-w-case-915-433mhz
 

Buzby

Senior Member
Hi hippy,

I fully understand the delay to ignore echoes, but if I need a range capability of from, say, 2m to 20m, and I have an echo from a wardrobe at maybe 8m, then I need to delay 8m to ignore echo. Now I can't measure down to 2m !

Indoor measurements are going to be full of spurious echos. Even the floor may echo before the mobile sees the trigger.

This is why I'm now convinced the Marvelmind is using RF to trigger the US.

Cheers,

Buzby
 

hippy

Technical Support
Staff member
I fully understand the delay to ignore echoes, but if I need a range capability of from, say, 2m to 20m, and I have an echo from a wardrobe at maybe 8m, then I need to delay 8m to ignore echo. Now I can't measure down to 2m !
I don't see why not.

You send a trigger and ignore everything for a second. The receiver sees the trigger, also delays a second and sends its echo back.

The time it takes for that echo to come from target to base station is the time after triggering minus one second.
Code:
              .--- 1 second ---.
              :                :
Tx Send    ___|_______________________
Tx Echo    _____|_|_|______________|__
               reflections     :   :
                               |<->|
RX receive ____|_|_|_|________________
Rx Send    _____________________|_____
               :                :
               `--- 1 second ---'
You can try it with a couple of friends. We'll go up to minutes with your target responding one minute plus one second per meter they are away from you. Your other friends are reflections. Blindfold on, stopwatch at the ready ...

00:00:00 Yoo hoo
00:00:01 Reflection
00:00:04 Reflection
00:00:08 Reflection
00:00:15 This is boring
00:00:20 I'm outta here
00:01:01 Hi there!

So 1 minute 1 second after your trigger you got back the response you were hoping for. Subtract the minute; leaving 1 second. So your friend is standing 1 metre from you.
 
Last edited:

Buzby

Senior Member
I've not given up on this, just been too busy with work.

I did get the ping-delay-reply mechanism working, but measurements were very erratic.

Another chance to try next week, hopefully I'll find the cause.

Cheers,

Buzby
 
Top