HB100 Radar speed detector

What are you developing?
User avatar
RogerClark
Posts: 7436
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: HB100 Radar speed detector

Post by RogerClark » Fri Aug 25, 2017 12:02 pm

Andy

I have not had chance to wire up my RCWL 9196 motion detector chip to the Doppler module.

But I tested the 2 RCWL 9196 modules that I had previous been playing with, and only 1 of the 2 is still working

I don't know if its the RF Osc that has failed or the RCWL 9196, I'm going to unsolder the RF Osc transistor from the defective board and see if the RCWL 9196 is still working.
But if not I can use the other RCWL 9196 that is still working, as I have one more of these boards, which I have not wired up so far.

User avatar
RogerClark
Posts: 7436
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: HB100 Radar speed detector

Post by RogerClark » Fri Aug 25, 2017 11:14 pm

It looks like the oscillator has failed on one of my motion detector boards, but the RCWL 9196, still seems to work.

However, when I tried to connect the IF output from the HB100 into the RCWL 9196, although it amplifies, it seems to have a minimum threshold which means that its only amplifying the signal if it gets quite strong.

I don't think its worth the hassle of attempting to modify the resistor and capacitor values on this tiny surface mount board

So, I'll get hold of a LM324 if they have one in stock at the local electronics shop, and build the circuit as recommended in the application data sheet

User avatar
RogerClark
Posts: 7436
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: HB100 Radar speed detector

Post by RogerClark » Sat Aug 26, 2017 6:54 am

I built the circuit from the application note, but it does not seem to perform as I would expect

Image

The output voltage settles at 2.5V when there is no movement, with about 1V of noise.

If I move my hand over the sensor, the output gets driven between 0V and 4V

Ideally when there is no significant movement, I'd rather the output stayed at either 0V or 4V, rather than floating in between

As the LM324 is a quad opamp, I could take the output from the current circuit and feed it into another (3rd) opamp in the same package e.g. feed it into the inverting input, and then bias the non-inverting input to around 1V, which should generally drive the output of a the 3rd opamp to 0V when the output is floating

User avatar
RogerClark
Posts: 7436
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: HB100 Radar speed detector

Post by RogerClark » Sat Aug 26, 2017 7:12 am

I found this

https://medium.com/@benjamindavidfraser ... 2d8cacf82e

He used a additional comparator stage

Image

User avatar
Pito
Posts: 1626
Joined: Sat Mar 26, 2016 3:26 pm
Location: Rapa Nui

Re: HB100 Radar speed detector

Post by Pito » Sat Aug 26, 2017 7:51 am

The output voltage settles at 2.5V when there is no movement, with about 1V of noise.
Try to add 100nF-1uF in parallel to first R3=12k (your schematics in previous post).
..you have to see DC 2.5V (plus minus opamp's input noise multiplied by amplifier gain, maybe ~hundred(s) of mVpp) at the output of the second opamp..
Noise: LM324 could be noisy at such gains (it is rather old general purpose opamp).
Is the noise of higher frequencies?
The best way to limit noise is to decrease the Bandwidth of the amplifier. In case you do not expect fast changes in the movement simply do lower the low-pass corner by increasing the feedback capacitors from 2.2nF to say 10nF or more.

R1/R2 virtual ground must be blocked ie 100uF || 100nF
Last edited by Pito on Sat Aug 26, 2017 8:05 am, edited 1 time in total.
Pukao Hats Cleaning Services Ltd.

User avatar
RogerClark
Posts: 7436
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: HB100 Radar speed detector

Post by RogerClark » Sat Aug 26, 2017 8:01 am

@pito

OK.

I'll add some extra decoupling to the virtual 2.5V and try adding a cap in parallel with R3

User avatar
Pito
Posts: 1626
Joined: Sat Mar 26, 2016 3:26 pm
Location: Rapa Nui

Re: HB100 Radar speed detector

Post by Pito » Sat Aug 26, 2017 8:12 am

As written in my previous post you may decrease the noise by increasing the 2.2nF feedback caps to 10nF or more.
With larger feedback capacities you will see just slow fluctuations around 2.5V (virtual ground) at the output (also the changes due to moving objects will be slower).

The noise will be always there. When you want to mess with speeds measurements or something like that you must DSP the signal somehow..

Or, use "low noise opamp", there are zillion types out there.

Another option is to use single transistor amplif as the first stage (usually much lower noise).

Another option is to lower the Gain of the opamps (decreasing the 1Meg to 100k-330k for example). Try, maybe the amplification will be enough.

If you want to have signal from zero to 4V you may try to use the third opamp like "opamp peak detector" or "opamp rectifier".

With using "comparator" at the output you will loose a lot of interesting information in the signal..

Of course the construction style of such high gain amplifier is important too (no ground loops, shielding, etc).
Last edited by Pito on Sat Aug 26, 2017 8:50 am, edited 5 times in total.
Pukao Hats Cleaning Services Ltd.

User avatar
RogerClark
Posts: 7436
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: HB100 Radar speed detector

Post by RogerClark » Sat Aug 26, 2017 8:21 am

As the LM324 has 4 opamps in the package,

I've wired up another of the opamps without any feedback, and have the wiper of a 10k trim pot on the non-inverting input and the output from the previous output (from the 2nd opamp) into the inverting input

(basically the same as that other circuit, except I'm using a 10k pot and an Opamp stage rather than a comparator)

This seems to work quite well, as it cleans up the signal into Highs and Lows (4V and 0V)

I'll need to add a resistor divider on the output so I can feed it into a Blue pill and attempt some pulse width analysts

User avatar
RogerClark
Posts: 7436
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: HB100 Radar speed detector

Post by RogerClark » Sat Aug 26, 2017 11:37 am

Just in case anyone else e.g Andy is interested....

I found this interesting article

http://physicsopenlab.org/2016/05/30/mi ... er-sensor/

User avatar
ahull
Posts: 1650
Joined: Mon Apr 27, 2015 11:04 pm
Location: Sunny Scotland
Contact:

Re: HB100 Radar speed detector

Post by ahull » Sat Aug 26, 2017 12:33 pm

RogerClark wrote:
Sat Aug 26, 2017 11:37 am
Just in case anyone else e.g Andy is interested....

I found this interesting article

http://physicsopenlab.org/2016/05/30/mi ... er-sensor/
Thanks. I'll add that to my to do list. Not sure when I'll get a chance to build it, as I am in the process of moving house at the moment.
- Andy Hull -

Post Reply