TMS37157: problem at low range

Did you trim the tag IC after changing the coil?

Thank you for your fast response.

Yes. Auto trimming was successful. (No red led flashing after procedure)

Johan

just to confirm -you trimmed it before you made the post, so trimming the part is not the issue...is that correct?

do you have any pix of the coil and your entire setup?

Hello Johan.

I suspect the Q factor of your "new coil" is too high (>>60?)

If the Q is too high, the OOK of the request for "Page 3 will being badly demodulated and even corrupted...

During the Transmitter encoding it gets turned/modulated on/off. During the "off" gaps, the Resonance artifact in the Transponder will not let the received OOK video envelope fall down to the the lower logic threshold decision point

Therefore for the 15 mS duration of the "Write Request" the tag never sees a meaningful 1>0 transition...or when very close they will blurr into one steady state logic 1 and be contigusious to the 25 mS Charge burst.

Your symptons seem to suggest that as you get closer the Tag begings to "think' it is seeing only one "long continusious Charge burst"

Apparently the tag is designed to "kill itself" if no (or a corrupt?) Write Request is received...the tag empties it's energy storage and prepares for a new "clean restart"...not sure but I think the Tag emits the 16 Pre-bits then discharges it self...a "last gasp?"

Your Oscilloscope view of the OOP modulation "show me Page 3"...in the last 15 mS of the Readers interrogation, seems to show oscillatory overshoot or "ringing" this being symptomatic of an under-damped resonance...ie the Q factor is too high and needs a new damping resistor...probably about 4.7K try to start with?

Also note, the next critical time line event is/would be the identification of the first "bit" of the Start byte....a very high Q will intrude into it and confuse the FSK decoder in the Reader.

I be interested what happens when you "re-dampen" your new coil.

Kind regards
Ray

Hi Josh,

I trimmed it before I made the first post.

I can send you pictures by email, but I am not allowed to post them on a public forum.

best regards,

Johan

got it - send over to : josh.wyatt@ti.com

Hi Ray,

The coil in the tag is near a coin cell battery. As a result, the Q is rather low (+-20). Coupling with the primary on the other hand is quite good (k=0.02).

If the Resonance artifact in the Transponder will not let the received OOK video envelope fall down to the the lower logic threshold decision point, then the EOB signal would not be correct, isn't it?

I tried to add 4.7k in parallel with the tag coil, but then I have no more range.

One difference I noticed between  a situation that works, and one that doesn't: (see first 2 pictures): In the situation where the tag responds, there is some oscillation of the tag coil when the reader coil is off (seems to be transmitted by the tag) . In the situation where the tag does not respond,this oscillation isn't there.

best regards,

Johan

My mistake Johan... I thought the EOB waveform was the "modulation" into the Reader's transmitter... I was not sure, (a little ignorance by me)... but now I see the EOB waveform is in fact the "Demodulated" video in the TMS37157... so yes my proposition about the "off" gaps disappearing/merging is clearly wrong. My apology.

However, I also tried to optically measure the On and Off times off your images and perhaps I saw that there was some changes happening as the k factor increased...in which case perhaps the deciphered address was "borderline unintelligible" for the TMS37157 and it decided as an invalid address it would ignore it and prematurely shut itself off?

There seems to be some critical numbers for the on off durations...my thoughts on detuning or "double-humping" as two resonant circuits as brought together look to be wrong...I will reconsider... so is the Battery purposely placed to "dampen" the resonance?

Your observation about faint "strange oscillations" after TX switch-off's remains my argument core...are you sure your before and after images are not transposed, the un readable one seems to be "cleaner" in the gaps.

Did 4.7KOhm completely stop detection (no more range) or just desensitize the process?... me still curious... try 10K-20K ohm then.

Regards

Ray

Ray -

he is using the TMS3705, this is LF, not HF (so not any TRF79xx in the picture here)

Johan - i think your coil might be too low Q, therefore this is perhaps the issue - have you tried using lower resistance coil to raise the Q factor? (i.e. fatter magent wire or litz wire coil)

Also, on the TMS3705 basestation board - you can remove diodes D2 and D3 and move R34 to position R35 and power with 12VDC to raise the output (downlink) voltage up.

See schematics for that here (at the end of the doc) ==> http://www.ti.com/litv/pdf/slau281b

Thanks Josh... me by accident referenced TRF79xx.. (fingerproblem here).

Johan,  please confirm your detection "ceases" as the Tag gets too close, ie <2.5cm

Josh your strategy seems to be solving an opposite problem, you are trying to arrange to harvest more energy by asking for a "higher Q" ?

In addition you recommend for "more" emitted charge field energy from the basestation board...ie use 12V etc...Johan seems happy with 7cm?

Johan, are the two coils facing each other as your TMS37157 device approaches (face-on or side-on?) your "custom built reader and or the original TI TMS3705 PCB. and how did you come by your coefficient of coupling figure, was it based on the ratio of coil apertures?

As you suggest, the problem does follow the "new coil"... therefore is it not the issue... and yet all it has is Aperture/Shape//Q/(DC and AC resistance) stray capacity and companion objects....choose one!

Josh, if the Q is to low as you propose ( and Johan does say about 20?), then what is the deeper reason for the VCC being prematurely dumped/short circuited?

Am I correct that these RFID Tags remain mute until they are asked (polled) to speak...unlike Cattle Tags...

Regards

Ray

Johan and Josh...is there a "look-up" table inside the TMS37157 that recognizes a "Valid" [Command+Pagefield] combination? ... ie spanning 00h to FFh

If a combination is found to be "not in this Valid list"...what does the TMS37157 do?

Am I correct that at least one situation is documented... that is... "after a charge phase only (any length?) see page 21 ... the tag dies prematurely ... so is this also the case for ANY "invalid" Hex code?

I am thinking any corruption of the on/off time durations defining a High or Low (or a spurious noise spike changing any bit perhaps) could force a genuine valid code into a death wish non-valid list and onto the symptom Johan is experiencing.

I think I can see a case for corruption from a low Q and a High Q TMS37157....it comes and goes ...more though/help needed...I keep thinking this is another species of the famous TI "holes"

Regards

Ray

Hi,

Regarding timings: on and off times change a little as the transponder gets closer, because the detection threshold is crossed sooner/later, but according to figure 4 of the TMS37157, what matters is the timing between two rising edges of EOB. "If the Pause between to positive transitions of EOB is at least as long as tHdet the Transponder writes a one. Is the Pause shorter it writes a 0." tHdet is 470 us, so the timings should be met.

I can confirm that the first image is the situation where the tag answers. In the second one there is no answer. The oscillations in the top figure, when the base station is off, seem to be actively transmitted. They have some sharp edges, as seen during FSK response.

Is there some state diagram of the TMS37157?

thanks again,

Johan

Johan, I put it to you that the Regenerated clock period is Xtal stable during Thigh and "significantly Unstable" during Tlow...after critical coupling (closeness) is exceeded.

I propose that beyond "critical coupling" of two resonant circuits there is bifurcation and now the "two resonance peaks" exist....which one does the clock regenerate to?

The self-made clock is now free-wheeling" in between the Huge PowerField bursts (which are Xtal locked to 134.2Khz)... =stable clock periods of 7.4516' uS counting....brute force regenerates a predictable clock for the first part ...but not for long.

The Clock Regenerator "plucking/recursive" creation during the "Transmitter Off times" now becomes indecisive...first part =stable good ...second part= unstable variable.

The Video modulation is still substantially the same but now the de-terminator of what represents a Binary 1 or 0 which "I think is a clock period counter 510/7.4516?"...

I suggest the clock period counter either ends of finishes too early or too late to declare it found a one or zero....sometimes a bad single "bit" is all it takes?...indeterminate decision based on physical distance? = your symptom?

Will reflect some more

Regards

Ray

Johan...

Battery placement and Device Docking Cradle are  to blame for all this.?

You are Data logging = why else is battery there in first place?

Battery too close to RFID coil ? = you say lowers Q ...yes, like a shorted turn you reason?...is not 20 still too high?

Battery too close to RFID coil =  me say minute charge/discharge battery currents magnetically coupled back into to Reader coil

Magnetic feedback creates "strange forced damped oscillations" in between the Reader transmit gaps = your top picture = plucking plus feedback? = superposition

Close proximity of Reader coil curiously/predictably  "dampens/absorbs" this spurious feedback energy...still a vestige = your bottom picture

Why you go to 24mm diameter coil = k factor between both coils now bigger over original small PCB inductor instance.

Why you go to 24mm coil= surely not for more range, SMD inductor is adequate for 2.5-7.5 cm ?...

Why you go to 24mm coil = is your device like an electric tooth brush which gets "docked then manually decanted?..seemed like a good idea at the time? 

Your Reader is not endlessly Polling otherwise your device would be successfully read and decanted as it came closer....and you be non the wiser about <2.5cm range?

Has any one else experienced what you are recounting here?

When you enacted the "internal Tag tuning" was the tag well away from the reader?...but close enough to get clocking...on the verge?

Me still think Clock definition confusion due to bifurcation is still a main item here see above.

Could you move the battery and go back to small aperture SMD ferrite coil?...the physical form of your device maybe brought all this on?

An interesting problem here...if it eventually though turns out to be a strange firmware matter, I will be surprised.

Regards

Ray

Hi Ray, Josh,

I just got a brand new EZ430-TMS37157 evaluation kit.

It shows the same behaviour: if the tag is too close to the base station (see photograph), the tag stops transmitting.

Maybe the problem is not related to the Q, only to the coupling factor that becomes too high?

best regards,

Johan