How to boost RFID signal

Thread Starter

rjjenkins

Joined Apr 16, 2011
177
Hello

I have an application that uses a 125 KHz RFID reader, a 10cm square antenna at 700 uH and passive nail-type RFID tags. The tags are embedded in objects that pass through the antenna. About 98 per cent of the time, the tag is read correctly, but 2 per cent of the time it is not read at all. It may be connected with the orientation of the tag relative to the antenna, which I cannot change.

The modules I am using are these

https://www.electrodragon.com/product/125khz-rfid-long-distance-module-40cm-serial/

I have tried all the suggestions listed on the site's wiki, such as smoothing the DC input with a capacitor and keeping the leads from the antenna to the module short. At the moment I have two ideas - one is to build in redundancy with a second antenna and module so that I get two chances to read the tag. The other is to boost the signal from the antenna. But I do not know whether this is possible, or whether it is simply a signal:noise problem.

Any thoughts?
 

Janis59

Joined Aug 21, 2017
1,232
Just idea rapid - bit make an antenna frame geometry bad, saddle-shaped. Then overall sensitivity will decreased by few percents, but skew-read sensitivity will grow for several tens %. Probably.
May look for Israeli Haifa Golden Youth publication they wrote in details how they built a frame able to check what RFID content is on street walker pocket if look from second floor window. After reading this article I made for my bank card an Faraday cage envelope in the pocket, because if Iraeli had only positive pure-experimental goal, then russian mafia here may have very most black reasoning.
The article leithmotive was standard cheap module what are boosted by FET to transmit and simple LNA boosted to receive. As well the impressive sized frame.
 

andrewmm

Joined Feb 25, 2011
632
there are two answers to this question, and you can do either or both,
a) Antenna , to give more directivity and hence gain, and hence longer range.
b) amplification,

Read up on antenna design , a Yagi would be something like 70m long, but other options are possible,
may be a ferrite antenna , as used in am radios,

Amplifiers, are relatively easy at these frequencies, many many companies out there, just do a search
 

Irving

Joined Jan 30, 2016
1,056
The difficulty here is knowing whether the failure is due to the tag not getting enough signal to power up, or the receiver not being sensitive enough to hear the tag, or both. Increasing the signal in the transmit antenna may resolve the first, while improving receiver sensitivity may resolve the second. Tag orientation will have a bearing on both possibilities.

If the problem is tag orientation I'd look at running another antenna in a different orientation, say at right-angles to the first. It needs to be linked to the first so they transmit simultaneously, and multiple receivers. Given 125kHz RFID is officially 10cm-ish range, 40cm or more is speculative under the best scenarios.
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
177
What confuses me here is how you would use an amplifier. The antenna sends out a signal, but it also receives back a signal from the tag. I presume it can't do both at the same time. How can you configure an amplifier so that it amplifies in each direction at the right time? Or would it be necessary to use two antennas?
 

Irving

Joined Jan 30, 2016
1,056
What confuses me here is how you would use an amplifier. The antenna sends out a signal, but it also receives back a signal from the tag. I presume it can't do both at the same time. How can you configure an amplifier so that it amplifies in each direction at the right time? Or would it be necessary to use two antennas?
There are various RFID protocols but all rely on a transmit period followed by a listen period. Boosting the output power with an amplifier into the antenna improves the chance of the tag responding but does nothing to improve the chances of hearing it respond. Making the receiver more sensitive improves the latter but then makes the receive chain more susceptible to blocking due to the transmitter overpowering the receiver front end so it takes too long to recover after the transmission. There are many ways to address that, though retrofitting it to an existing solution may be hard; two antenna is certainly one option.
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
177
Thanks. There seems to be no harm in trying to boost the output. At the moment, the scope shows a peak-to-peak voltage of about 60V. What sort of higher voltage would be better, and what kind of amplifier should I be looking for?

I tried a second antenna in the same plane as the first one, with a second receiver, (with an arduino using two separate serial inputs to detect either one presenting data). The problem seemed to be that the two antennae interfered with each other unless they were a few cm apart, which is a problem for my application. I tried putting one antenna perpendicular to the first, which helped when the tag was near it, but that still leaves a blind spot opposite the second antenna, and if I go for a third I will start running out of serial inputs (plus my attempts to wind my own antennae using formulae online have produced disappointing results.) I will try winding an antenna on a ferrite core but I suspect that even if it works it will be too directional.
 

Irving

Joined Jan 30, 2016
1,056
Directionality may actually help. A ferrite antenna rod will be far more sensitive axially, and at these low frequencies will have a fairly wide aperture. Could be worth trying.

60v is quite high, do you know what the current, or the expected output power, is? Hard to say what amplifier you need without knowin more about the current circuitry. My gut feel is you need to have separate transmit and receive antenna and the problem is receiver sensitivity rather than transmit power... You could test that by putting the receiver near the tag while moving the transmitter away, assuming you can split it easily that way.
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
177
I tried winding a ferrite rod to what should have been 700uH, but got a pretty rough looking waveform, low amplitude and poor reading range.
Unfortunately I can't separate transmitter and receiver - I'm using an off the shelf Chinese module that just has a couple of terminals for a single antenna. And no datasheet.
I'd happily build my own if I thought I could get a better range but I don't really know where to start. (The reason I'm using 125KHz is that I understand there is less attenuation than with the higher frequency systems, and the nail-type tags are cheap and robust.)
 
Hi rjjenkins,

I couldn’t open the link, but I have experimented with 125KHz RFID and developed my own reader with a TI chip and Microchip MCU.

In any event, increasing the current in the antenna should help with communication. The following circuit idea might be adapted:

https://www.electronicdesign.com/te...cuit-generates-a-highfrequency-magnetic-field

Typical RFID modules use a square wave driver and series resonance to create the sinusoidal high-current in the antenna. A probable issue with the circuit idea is that it requires a sinewave source. I’m sure the higher frequencies in a square wave driver will see the series caps in the circuit as a short. This may be overcome with an impedance matching inductor (L-match circuit) in series with resonant circuit, but I’m not certain how to calculate this. Perhaps another member may know to do this.

Good luck! I wish you success and found that rolling my own RFID to be challenging and fun!
Spitsnsparkin.
 

Irving

Joined Jan 30, 2016
1,056
I tried winding a ferrite rod to what should have been 700uH, but got a pretty rough looking waveform, low amplitude and poor reading range.
Unfortunately I can't separate transmitter and receiver - I'm using an off the shelf Chinese module that just has a couple of terminals for a single antenna. And no datasheet.
I'd happily build my own if I thought I could get a better range but I don't really know where to start. (The reason I'm using 125KHz is that I understand there is less attenuation than with the higher frequency systems, and the nail-type tags are cheap and robust.)
Hmmm, no information, schematic, etc at all? Can you take some hi-res pics of both sides of PCB and note any chip IDs, maybe we can figure it out?
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
177
It looks very like the EM4095 chip but one vendor of the module says that it "far exceeds" the performance of that chip. When I have a moment I will see if the pinouts correspond at all.
 
Hi rjjenkins,

I have to make a correction, I used an NXP HTRC1100 with a Microchip MCU.

My goal wasn’t distance, but reading a transponder implanted in the gut of rodents. The water in the body greatly reduced the signal.

I did in fact, split my antenna into two equal-turn loops in series. I over-wound both coils, then reduced the turns while dip-checking with a function generator and tank capacitor on a breadboard until the inductance was tuned to 125KHz. They were placed in front and behind the animal as the center of their body needed to be open. I don’t know if that improved the measurement range, it was simply a mechanical design necessity.

If the modules you have use an EM4095 type PLL circuit, that might explain why your two (module and antenna) test didn’t work. A beat frequency was probably created between the two. If the modules are crystal based, then it’s possible to use its driver output to control an external mosfet half-bridge with larger voltages to get more current and consequently a larger signal for the receiver.
However, I found that the EM4095 and HTRC1100 chips have a limited voltage range to their receiver inputs. As a consequence a series resistor could be added to the driver to reduce the Q and voltage of the antenna if it was too large. Similarly, a capacitor divider was placed across the tank capacitor to reduce the voltage input for the receiver. Therefore, increasing the antenna current beyond a point will require you to also reduce the antenna signal as not to exceed the input voltage of the receiver circuit, kind of self-defeating.
If the module is based on an HTRC1100 type, then a calibration procedure to the phase-adjustable sample-and-hold is greatly advantageous to getting the best signal.
Spitsnsparkin.
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
177
Thanks. I do not think there is a crystal.

When I tried two modules, I found that while one of them was spot on 125kHz, the other was about 122kHz, which would certainly have caused beats. Even if they were exactly the same, the inevitable phase difference would have reduce the amplitude, I'd have thought. In fact, however, I orient the two antennae, the result seems to be worse than having a single antenna.

I've been wondering if I split the single 700uH antenna into four separate ferrite antenna (with a combined inductance of 700uH) and positioned them to give best coverage of the reading area, that might work.

Alternatively there is a slightly lower frequency standard for tagging animals and there are modules that claim to give a slightly better range, which I can investigate.
 
Hi rjjenkins,
Personally, I don’t think ferrite cores In the antenna will enhance distance. As the working principle of the RFID topology is an air-core resonant transformer (to pass power to the tag and then read data from it by how hard it “loads” the primary). If the passive tags do not have a ferrite core within its antenna loop, I don’t see efficient coupling. The implantable 125KHz RFID tags had a small ferrite core antenna encapsulated in a glass ampule. When placed within the field, the tag’s ferrite core became a core for both antennas (reader and tag).
Spitsnsparkin.
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
177
I'm not sure that distance is my exact problem. The tags are read almost every time, often from a good distance (eg 8 cm). Just occasionally they fail, and it's hard to detect a pattern, unless it's that there are small blind spots in the field. I thought that several antennae spaced around the reading area might produce a better field, but I'm clutching at straws here.

I think it's possible that the nail-type tags do have a ferrite core though I've never opened one up. I'm planning to try the glass ampoule animal ones which are about 30mm long and so possibly the same as the ones you used.
 
Top