Bluetooth module question

[icy]

I'm using the Bluetooth Mate Silver from Sparkfun and I'm trying to measure the power of the bluetooth signal it receives using an RF detector. The received signal has to be input to the RF detector, which outputs a voltage proportional to power. I was wondering if it's correct to assume that the received signal can be measured at the RX pin, so I can just connect the RX pin and Ground to the input of the RF detector. It seemed obvious but now I'm not sure! Thanks for any ideas!

[Soeren]

Hi,

The received signal has to be input to the RF detector, which outputs a voltage proportional to power.

Isn't it supposed to measure on a continuous wave?

I was wondering if it's correct to assume that the received signal can be measured at the RX pin, so I can just connect the RX pin and Ground to the input of the RF detector. It seemed obvious but now I'm not sure!

The RX pin is a demodulated signal going to your microcontroller and has no measure of HF power.

Why measure at all?
If it's for comparison, just do an A/B test to see which gives a greater range.

[icy]

I know the RX signal has no measure of power--it's the RF detector I'm using that outputs a power measurement. The RF detector requires an RF input, and so I'm trying to understand what to connect from the Bluetooth module to the RF input.

This is the RF detector I'm using:
http://www.linear.com/product/LTC5531

Thank you!

[Soeren]

Hi,

I know the RX signal has no measure of power--

Oh, the sentence:

[...] I was wondering if it's correct to assume that the received signal can be measured at the RX pin, so I can just connect the RX pin and Ground to the input of the RF detector.

fooled me  

it's the RF detector I'm using that outputs a power measurement. The RF detector requires an RF input, and so I'm trying to understand what to connect from the Bluetooth module to the RF input.

What I meant was... You cannot get a thing from measuring the RX pin. It's a logic level signal that is not capable of giving any info to the LTC5531.

For this you need to connect your add-on at a point before the demodulation, with the risk of upsetting the entire circuit when measuring and thus getting no or a false reading.

Even if you succeed in tapping the signal without any impact (highly unlikely!), I don't tink that chip will be kind to your sleeping pattern because...
Your Bluetooth module is described as short range, so I'll assume a class II device. This means a maximum output of 4dBm.

At a 10m distance, the free space attenuation will be around 60dB and around 40 dB at 1m distance. That is with nothing but free air and no obstacles in between.

So, the strength at the receiving antenna will be somewhere between -36dBm and -56dBm at a 1..10m distance.

I have no idea how much gain the receiver stage provides at HF, but I would be very surprised if you could find a signal with enough amplitude to actually get any usable info out of the LTC5531 that have a lower limit of -32dBm.

Usually, you couple such devices with something called a weak loop, to avoid quenching the circuit that you want to test, but that will introduce a huge loss more.


So, I'm back at the still unanswered "Why measure at all?"

[icy]

Thanks for the insight. I just want to create a Bluetooth signal strength indicator as a project.

[Soeren]

Hi,

Thanks for the insight. I just want to create a Bluetooth signal strength indicator as a project.

If it's just for some school project (or similar), I'd go at it with a field strength meter instead.


I'd add a resonance circuit at the input to get a little more oomph at the frequency of interest and a little less at other frequencies.

This will only work at close range, but if it's just for checking/comparing, it will be fine.
For added sensitivity a low noise microwave transistor (like BFU725F with a gain of 25.5dB at 2.4GHz) can be used to up the HF signal (i.e. to the left of the diode demodulator).

[icy]

Thanks for the suggestion--I'm thinking of a simpler method and I was wondering if you have insight about this. I have a demo board for the LTC553 RF power detector that has an RF input and Vout pin that gives the voltage proportional to power. I connected a 2.4 GHz antenna (from http://www.sparkfun.com/products/145) to the RF input, and the Vout pin to the ADC of a microcontroller, with the intent of viewing the ADC output on an LCD screen that is interfaced with the microcontroller. I have bluetooth on my computer so I thought the antenna could pick up the signal. However, nothing seems to be happening and I don't know how to troubleshoot!  

[Soeren]

Hi,

Thanks for the suggestion--I'm thinking of a simpler method [...]

Don't ask if you can't handle the answers
A field strength meter is the simplest method!

[...] and I was wondering if you have insight about this. I have a demo board for the LTC553 RF power detector that has an RF input and Vout pin that gives the voltage proportional to power. I connected a 2.4 GHz antenna (from http://www.sparkfun.com/products/145) to the RF input, and the Vout pin to the ADC of a microcontroller, with the intent of viewing the ADC output on an LCD screen that is interfaced with the microcontroller. I have bluetooth on my computer so I thought the antenna could pick up the signal. However, nothing seems to be happening and I don't know how to troubleshoot!  

Please read my previous answers and you'll know that it won't work that way - nothing to troubleshoot, you simply haven't got enough signal for this and nothing will change that fact.

Either make a field strength meter, or make a complete 2.4GHz receiver that can act as a preamplifier stage for the chip.

[icy]

Thanks for the info! Is the option of creating "complete 2.4GHz receiver that can act as a preamplifier stage" a considerable task for a student/hobbyist? I'm having trouble finding information about how to do this. Looking up this topic online, I am only finding preamplifier products for TVs!

[Soeren]

Hi,

Is the option of creating "complete 2.4GHz receiver that can act as a preamplifier stage" a considerable task for a student/hobbyist?

Yes, I know a good many engineers that wouldn't take it on.
Besides, building it for the purpose would be crazy, as a field strength meter is a very easy thing to make. Put your chip aside and build a simple meter, it will do the job with minimum headache

I'm having trouble finding information about how to do this. Looking up this topic online, I am only finding preamplifier products for TVs!

Build a field strength meter!
The time you save by following that route will be available for the rest of your project

[icy]

I finally got the parts to build the field strength meter...and I can't get it to work! I used the 2.4 GHz duck antenna from  http://www.sparkfun.com/products/145 and got a reverse polarized SMA to BNC adaptor so that the antenna can connect to a bnc cable with alligator clips on the other end. I connected the alligator clips across the nodes shown in the schematic. Do you see anything wrong with this?
Thanks...

[Soeren]

Hi,

I finally got the parts to build the field strength meter...and I can't get it to work! I used the 2.4 GHz duck antenna from  http://www.sparkfun.com/products/145 and got a reverse polarized SMA to BNC adaptor so that the antenna can connect to a bnc cable with alligator clips on the other end. I connected the alligator clips across the nodes shown in the schematic. Do you see anything wrong with this?

Yes, a field strength meter is a simple handheld unit with a whip (or a coil) mounted directly on the box. As soon as you introduce alligator clips, you mismatch the antenna completely.

One common mistake in making HF circuitry is the practical layout. At 2.4GHz, anything is a component(*), so stay with a 29mm copper wire mounted directly at the input and make sure it connects to the 51 Ohm resistor and the two diodes with minimal wiring (i.e. make it a single pad where they all connects). At the right side of the diode (in the schematic) you needn't worry over high frequencies.

If it still won't work, try posting some sharply focused photos of your circuit.

When measuring, you need to place the meter very close to the transmitter you want to measure, as there isn't much signal to begin with.
If you want to compare different BT units, make sure the antenna whip is oriented parallel to the antenna inside the BT and at the same distance.


*)  When laying out PCB's for GHz frequencies, I use a micro-wave CAD program, as this takes into account the PCB material, the thickness and dielectric constant of the PCB carrier material and the thickness of the copper foils (all parameters must be entered beforehand of course) and can thus calculate the length/width for stripline traces, capacitors and inductors. Such a PCB cannot be cloned by anyone without pretty good micro wave skills, as less than 1/10mm difference in the traces will alter the frequencies too much. The upside is that most of your L's and C's are just traces on the PCB, so in that way it's cheap and precise.

[icy]

There's something I'm not understanding about the schematic. If the copper wire is the antenna, then it looks like the copper wire is in parallel with the resistor, shorting it out...and what does the Ground symbol represent, because there's no battery?
Also what did you mean when you wrote "At the right side of the diode (in the schematic) you needn't worry over high frequencies"? Not to include this part of the circuit?
 
Thanks for your help!

[Soeren]

Hi,

There's something I'm not understanding about the schematic. If the copper wire is the antenna, then it looks like the copper wire is in parallel with the resistor, shorting it out...and what does the Ground symbol represent, because there's no battery?

Battery or not, the circuit need the common ground plane and since you need the amplifier, it does need a battery as well.
The (+)9V connected to pin 4 in the schematic assumes you're using the LM324, so get a datasheet for whatever chip you use.

Apart from the chip, the best way to make a prototype circuit like that is using "Manhattan Style", where you take a piece of unetched PCB, copper side up, glue little islands of PCB material where the physical nodes (solder pads) are and use component pins or *short* length of wire as interconnects.
You could combine it with Dead Bug mounting for the chip, or make IC-pads for Manhattan Style out of stripboard.


The input IS in parallel with the resistor, but it's not shortening it - it's impedance matching for the input, since it use only a whip (which is a high impedance) for your purpose, the tolerance doesn't have to be too tight (eg. 47 Ohm will be fine), but some radio amateurs connects it to their (tuned) antennas where it really ought to be 50 Ohm - Since it's only used as a relative indicator and does not transmit, it doesn't matter all that much there either.

You don't connect the antenna whip to ground, but the ground plane of the circuit will be used as the (artificial) ground plane for the circuit - if you need an explanation for that, read up on antenna theory.

This is about as simple as a field strength meter come. More sensitive and selective units can be build with a better front end, but I see so many people shy away from anything with coils, so I went with this one to make it less "scary".


As I mentioned earlier, HF and the higher ranges of frequencies in particular, is very different from DC circuits, so a common DC analysis will be useless.
To be able to deliver both a functioning circuit and more important, a satisfying documentation on your project, you need to brush up on both AC analysis and HF and it wouldn't hurt improving your schematics reading skills if you get some time left over.

Also what did you mean when you wrote "At the right side of the diode (in the schematic) you needn't worry over high frequencies"? Not to include this part of the circuit?

I meant that in the right side (from the diodes, towards the hand you use to shake hands with ) of the schematic, there's no high frequency to worry about, as the two diodes peak rectify the signal and the 2 caps (10pF and 1nF) integrates the signal to "varying DC" (only real DC with a constant amplitude signal of course).