• State TI Thinks Resolved
  • Locked Locked
  • Replies 19 replies
  • Answers 4 answers
  • Subscribers 19 subscribers
  • Views 2922 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TMS3705B used as receive only

Mauro Kerry
Prodigy 115 points
Other Parts Discussed in Thread: TMS3705, UCC27424

I am using a TMS3705B as a receiver in a system that generates the field with a separate transmitter.  i.e. the 3705 antenna drivers are not used or connected.  The data stream from the tag is incorrect in this arrangement.  If I use the built in driver transmitter then it works correctly.  Is there a fix for this?

  • 0
    TI__Mastermind 26420 points
    Hey Mauro,

    Are you using the same antenna for the 2 configurations?

    So you are saying that you are leaving the Full bridge un-connected, but are wiring up the rest? Could you provide that part of the schematic to help us understand?

    Thanks,
    JD
  • 0 in reply to JD Crutchfield
    TI__Guru* 91086 points

    Mauro - 

    if i understand your description correctly - you want to use the TMS3705 as a sort of listener or sniffer - is that correct? 

    this may not be possible without lowering the TXCT line for at least 20mSec then raising it to kick off the receive phase inside the chip as the default mode of the part is a read-only mode that uses the default frequency as the carrier frequency for the full bridge. Therefore the mode control register does not need to be written (it is filled with low states), and the communication sequence between microcontroller and base station starts with TXCT being low for a fixed time to initiate the charge phase. When TXCT becomes high again, the module enters the read phase and the data transmission via the SCIO pin to the microcontroller starts (this would be the response from the tag).

    so - perhaps you can synchronize your other TX'er operations with the TXCT pin operation and make it work out - not sure - like JD said, would be helpful to understand what you do and don't have connected, too. 

     

  • 0
    Prodigy 115 points
    Thanks Josh for your prompt reply. 

     Let me elaborate a bit.  The tag system that I am working on is for fish tagging and is intended to detect both FDX and HDX tags.  It uses moderate power to the antenna loop.  Six watts or so.  The FDX portion is done and works pretty well.  For HDX I pulse the field 60/30ms as you normally would and slave the tcxt to the transmit on/off control line.  I really wanted to use the 3705 since I think it's digital demodulation is better than a PLL or FM receiver but couldn't get it to work.  It would output a data stream that was consistent but wrong.  e.g. 62C10400D60300B233F908 for 7EB458300180F50080EC  which is the correct code for that tag.  By the way, the correct code was read by the same 3705 that read the incorrect code by attaching an small antenna directly to the 3705 antenna driver circuits.  In our system the transmitter uses a separate, low noise, 134KHz signal generator to accommodate the FDX am tag.  I tried to use the 3705 for this but its phase noise was much too high.

    My guess is that the problem is that the 3705 needs all the pre-bits to sync to the incoming data train from the tag.  The data I am getting seems to indicate some sort of framing error.  My design uses moderately high Q circuits in both the transceiver and the antenna and it is difficult to quench the field completely in time to get all the pre-bits.  The scope indicates that I am missing the first 3 or 4 bits.  Any other ideas?

    Thanks again,

    Kerry



    On Wednesday, March 18, 2015 6:47 PM, Josh Wyatt <bounce-614@mail.e2e.ti.com> wrote:


     
    A Message from the TI E2E™ Community
    Texas Instruments
     
    		 Josh Wyatt
    Mauro - 
    if i understand your description correctly - you want to use the TMS3705 as a sort of listener or sniffer - is that correct? 
    this may not be possible without lowering the TXCT line for at least 20mSec then raising it to kick off the receive phase inside the chip as the default mode of the part is a read-only mode that uses the default frequency as the carrier frequency for the full bridge. Therefore the mode control register does not need to be written (it is filled with low states), and the communication sequence between microcontroller and base station starts with TXCT being low for a fixed time to initiate the charge phase. When TXCT becomes high again, the module enters the read phase and the data transmission via the SCIO pin to the microcontroller starts (this would be the response from the tag).
    so - perhaps you can synchronize your other TX'er operations with the TXCT pin operation and make it work out - not sure - like JD said, would be helpful to understand what you do and don't have connected, too. 
     
    Did this answer your question?
     
    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.

    Flag this post as spam/abuse.


  • 0 in reply to Mauro Kerry
    TI__Guru* 91086 points

    Kerry -

    the TMS3705 is designed to work with low Q antennas in the first place with a value of between 440uH and 700uH and the gain of the receiver should not exceed 5 (set by Rf and Ri resistors on SENSE and SFB pins)

    and the downlink to HDX tags is nominally 50mSec charge, then TX should be turned off to allow those tags to respond (in about 20mSec time)

    perhaps you are missing the first bits because of the natural filter you have in place with the  high Q antenna or the timing is off?  you said you are seeing with a scope and you can see what the tag is actually coming back with - can you share that?

    see attached for example of what would be expecting to check out. 

    7026.Notes on TMS3705A1DRG4 & Read.pdf

  • 0
    Prodigy 115 points
    Hi Josh,

    This is about the best that I can show you.  The time base is 100us/div. and though you can't see it in the picture the decay extends out to 5 divisions.  That means that there is a collision with the pre-bits and the first few are missed.  The tuned transformer at the transmitter is pretty beefy and stores a lot of energy in its air gap so I don't think there is a fix for this.  Maybe when your new part 4140? becomes available it would be worth another try.

    Thanks again,

    Kerry  



    On Thursday, March 19, 2015 10:56 AM, Josh Wyatt <bounce-614@mail.e2e.ti.com> wrote:


     
    A Message from the TI E2E™ Community
    Texas Instruments
     
    		 Josh Wyatt
    Kerry -
    the TMS3705 is designed to work with low Q antennas in the first place with a value of between 440uH and 700uH and the gain of the receiver should not exceed 5 (set by Rf and Ri resistors on SENSE and SFB pins)
    and the downlink to HDX tags is nominally 50mSec charge, then TX should be turned off to allow those tags to respond (in about 20mSec time)
    perhaps you are missing the first bits because of the natural filter you have in place with the  high Q antenna or the timing is off?  you said you are seeing with a scope and you can see what the tag is actually coming back with - can you share that?
    see attached for example of what would be expecting to check out. 
    Did this answer your question?
     
    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.

    Flag this post as spam/abuse.


  • 0 in reply to Mauro Kerry
    Expert 1130 points

    Hello Maura.

    Just some thoughts on you problem.

    Some simple Oscilloscope viewings on pin2 (SFD) will reveal classic and any odd analogue behavior.
    Related to HDX only reception. I suspect you have “inter-digit” corruption due to three things.

    In the RFID Cattle Industry, this business of allowing use of HDX together with FDX caused predictable chaos. Some manufactures were forced to do what you are undertaking. A box of tricks that equally well reads HDX or FDX.

    Australia drew a line in the sand and said “enough, enough ... of this madness” ...and mandated only HDX tags. Pretty wise move, however other (big) countries are still in state of confusion/denial. However, in the case of Fish tagging, FDX tags seem to be able to be physically smaller...FDX also claim to be cheaper. FM Capture-Effect is true secret to advantage of HDX. All that for another story.

    What your Oscilloscope “sees” on SFB (pin2) should help you decipher what is wrong with your HDX detection.

    What do you mean when you say your FDX part works “...pretty well...”? Good but not good enough? ... more! Is 90% Read efficiency good. (I think not?). Want same number for FDX and HDX surely?

    What do you mean by “... pretty beefy...” Why is it tuned? Just asking for trouble there? Why not use a no air gap torrid ferrite? ... another variable for you to play with? (see RFM007 circuit). Your circuit?

    Most likely, you have multiple design oversights: all related to your new distributed (HDX) Receiver and/or Transmitter resonances and Q factors.

    Though also you have the curious mentioned resonance in your HDX -Air Gapped-Transformer (bit ***/careless to do this? come back later)

    First, rebuild your original TMS3705 experimental board with known good small aerial. Out of curiosity, how big is your new satellite Loop-Aerial...

    More importantly why are you even doing this HDX-FDX separation concept; surely not just because .....??.
    Does your companion FDX also have its own equal dimensions Loop-Aerial? Are the satellite FDX and HDX aerials axially aligned so as to be equally fair to either tag? Just so many questions for you!

    Any how, maybe you are experimenting to create a Stream/Estuary monitor with one huge submerged Transmit Loop accompanied by many little FDX/HDX satellites (to address data collision issues of poor count performance ... perhaps I being too obscure just now? Let us continue.

    NOW use Oscilloscope and probe onto pin(2) (SFB), proceed to photograph a set of “known good reference waveforms”. Locate a tag at far/near/close positions. Be careful to avoid receiver saturation of far location tag, as telegram envelope shape here will best help diagnose your problem(s)?

    Do you switch select either FDX or HDX; Tell us, otherwise it adds another layer of distraction. That is, do you want to want to Read/Count passing swarms of co-mingled fish tags, or only swarms of homogeneous tags. Tell us. Me seeing more and more problems ahead for you. Big fish,tiny fish, coil aperture size, detection range, saltwater detuning ,tidal detuning, ...

    If you are endlessly firmware alternating between FDX-HDX-FDX-HDX-..... etc , have you EXTENDED the muting duration of the external FDX oscillator to be 60ms + 30mS (your numbers). Tell us.

    Now, deal only with HDX; kill/disable/remove entirely the FDX section. Remove one malicious variable?

    Coil Resonance should be mid way between 124 KHz and 134 KHz, (use a sweep generator if you have one?) else-wise lop-sided /different analogue amplitudes of different adjacent Bits. Oscilloscope will reveal this.[Retune resonating inductor or capacitors until ending few cycles of both digit types are equal ....now symmetrical either side of resonance peak = good thing for the following sine to square amplifier saturation conversion]

    Excessive Q factor causes longer than expected “foreign ringing” extending into the next digit. The TMS3705 waits/ignores first eight cycles, until it decides it is looking at a clean 16 cycle digit One or a 16 cycle digit Zero. (Sorry, I need data sheet clarification on this, me a bit hazy here ... I do not have a TMS3705)

    Suspect likely only several things need “tweaking”. Best you guide us(me), in small steps.

    Why not just show us your schematic? Any how ... and what is this business about of “item 4140” all about? Or is it secret? ... do you yet have an off forum answer?

    Bit strange, your new HDX detector produces, "from nowhere" Two extra bytes (F9 08) ? Should not there be another 7E after the first 7E, or is this not an animal tag you are demonstrating? (sorry me vague). So being consistent suggest "stray electrical noise pickup not a problem?

    e.g. 62C10400D60300B233F908 for 7EB458300180F50080EC = which is the correct code for that tag?
    Does the decoded faulty read, (62C.......908) change, as the tag approaches from infinity?

    Help us to deduce , by publishing/make public your Oscilloscope experimental views.

    Kind Regards

    Ray

  • 0
    Prodigy 115 points
    Hi Josh,

    Well, I got it to work.  If you look closely at the timing it turns out that the transmitter shuts down 1ms after TxCT goes high.  I'm guessing that the de-modulator needs that 1ms of 134KHz signal to get itself synced up to the incoming data frequency.  Seems like this would be handled internally but apparently not.  

    Kerry 



    On Thursday, March 19, 2015 6:29 PM, Kerry Mauro <kmauro2@yahoo.com> wrote:


    Hi Josh,

    This is about the best that I can show you.  The time base is 100us/div. and though you can't see it in the picture the decay extends out to 5 divisions.  That means that there is a collision with the pre-bits and the first few are missed.  The tuned transformer at the transmitter is pretty beefy and stores a lot of energy in its air gap so I don't think there is a fix for this.  Maybe when your new part 4140? becomes available it would be worth another try.

    Thanks again,

    Kerry  



    On Thursday, March 19, 2015 10:56 AM, Josh Wyatt <bounce-614@mail.e2e.ti.com> wrote:


     
    A Message from the TI E2E™ Community
    Texas Instruments
     
    		Josh Wyatt
    Kerry -
    the TMS3705 is designed to work with low Q antennas in the first place with a value of between 440uH and 700uH and the gain of the receiver should not exceed 5 (set by Rf and Ri resistors on SENSE and SFB pins)
    and the downlink to HDX tags is nominally 50mSec charge, then TX should be turned off to allow those tags to respond (in about 20mSec time)
    perhaps you are missing the first bits because of the natural filter you have in place with the  high Q antenna or the timing is off?  you said you are seeing with a scope and you can see what the tag is actually coming back with - can you share that?
    see attached for example of what would be expecting to check out. 
    Did this answer your question?
     
    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.

    Flag this post as spam/abuse.




  • 0
    Prodigy 115 points
    Hi Ray,

    Thanks for the reply.  The good news is that I discovered what was causing the data read problem for the 3705 and have now got it to work.  It turns out that the 3705 continues to transmit for 1ms after the TcXT line is taken high.  My guess is that it is using this time to set up an initial reference frequency for the digital demodulator.  That's just a guess however but by creating the 1ms overlap the problem went away.

    As for the rest the system I am currently working on is the result of starting out 10 years ago with an AllFlex reader and a Palm Pilot data logger.  Over the years it has evolved into a 2X3 ,6 channel, multiplexed system that uses a 75 ohm transmisson line between the reader/data logger and the antenna loops.  This has all been based on the AllFlex reader but due to supply problems I have had to design our own reader.  Didn't want to do that..  The design is now complete enough to go to production prototype with the solving of this last problem with the 3705.

    A few details which may address some of your comments.  The reason for the air gap is to control the Q of the tank circuit at the transmitter.  The 75 ohm line is linked coupled to the tank.   At the antenna end the 75 ohms is transformed back up to the antenna impedance.  The antenna is a single turn and is as high a Q as we can make it to generate as strong a field as we can get.  The single turn is very important when running in water.   Avoids all the dielectric problems of multi-turn antennas.  Transmission line feeding is very effective and allows the antenna to be many 100's of feet away from the electronics.  With 6-8 watts we get 4 m^2 of internal antenna area read with 12mm FDX tags.  

    I like the FM tag better that the AM tag.  Particularly when there is noise.  However, there is a lot of work with juvenile fish that requires the small 8mm tags.  Studies are now being done over entire watersheds and larger fish are often tagged with FM, HDX, tags for obvious reasons ,and that is why dual capability is desirable. 

    As long as TI still makes the part I will stick with it,  Once it is working it seems to be pretty trouble free.

    Kerry



    On Tuesday, March 24, 2015 7:38 AM, Mauro Kerry <bounce-3194868@mail.e2e.ti.com> wrote:


     
    A Message from the TI E2E™ Community
    Texas Instruments
     
    Hi Josh,

    Well, I got it to work.  If you look closely at the timing it turns out that the transmitter shuts down 1ms after TxCT goes high.  I'm guessing that the de-modulator needs that 1ms of 134KHz signal to get itself synced up to the incoming data frequency.  Seems like this would be handled internally but apparently not.  

    Kerry 



    On Thursday, March 19, 2015 6:29 PM, Kerry Mauro <kmauro2@yahoo.com> wrote:


    Hi Josh,

    This is about the best that I can show you.  The time base is 100us/div. and though you can't see it in the picture the decay extends out to 5 divisions.  That means that there is a collision with the pre-bits and the first few are missed.  The tuned transformer at the transmitter is pretty beefy and stores a lot of energy in its air gap so I don't think there is a fix for this.  Maybe when your new part 4140? becomes available it would be worth another try.

    Thanks again,

    Kerry  



    On Thursday, March 19, 2015 10:56 AM, Josh Wyatt <bounce-614@mail.e2e.ti.com> wrote:


     
    A Message from the TI E2E™ Community
    Texas Instruments
     
    		Josh Wyatt
    Kerry -
    the TMS3705 is designed to work with low Q antennas in the first place with a value of between 440uH and 700uH and the gain of the receiver should not exceed 5 (set by Rf and Ri resistors on SENSE and SFB pins)
    and the downlink to HDX tags is nominally 50mSec charge, then TX should be turned off to allow those tags to respond (in about 20mSec time)
    perhaps you are missing the first bits because of the natural filter you have in place with the  high Q antenna or the timing is off?  you said you are seeing with a scope and you can see what the tag is actually coming back with - can you share that?
    see attached for example of what would be expecting to check out. 
    Did this answer your question?
     
    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.

    Flag this post as spam/abuse.




     
     
    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.
    Flag this post as spam/abuse.


  • 0 in reply to Mauro Kerry
    TI__Guru* 91086 points
    Kerry -
    sounds like you have found the pre-bit phase (16 lows from transponder) which is nominally 1.9mSec, to allow the RF module to recover from the power burst.
  • 0
    Prodigy 115 points
    Actually not.  The field excitation continues for 1ms after TcXT goes high.  You can see this by watching the ant1 or ant2 signal and the TcXT line.  My guess is that this overlap is necessary for the receiver to establish an initial reference frequency for decoding.  You would think that it could be done internally but with the autotune maybe not.  Anyway, by shutting down my field generator 1ms after setting TxCT high fixed the problem.

    Kerry  



    On Wednesday, March 25, 2015 6:14 AM, Josh Wyatt <bounce-614@mail.e2e.ti.com> wrote:


     
    A Message from the TI E2E™ Community
    Texas Instruments
     
    		 Josh Wyatt
    Kerry -
    sounds like you have found the pre-bit phase (16 lows from transponder) which is nominally 1.9mSec, to allow the RF module to recover from the power burst.
    Did this answer your question?
     
    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.

    Flag this post as spam/abuse.


  • 0 in reply to Mauro Kerry
    TI__Guru* 91086 points
    are you still using high Q antenna and what you are seeing is actually ringing?
  • 0
    Prodigy 115 points
    No, I used a circuit identical to the one in the app note.  I just used the 3705 in a standalone configuration.  All by itself.  Then measured the signals ant1 & 2 and TcXT and saw what I described.  After this I went to my prototype system and set up the correct overlap and everything worked fine.

    Kerry



    On Wednesday, March 25, 2015 8:46 AM, Josh Wyatt <bounce-614@mail.e2e.ti.com> wrote:


     
    A Message from the TI E2E™ Community
    Texas Instruments
     
    are you still using high Q antenna and what you are seeing is actually ringing?
     
     
    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.
    Flag this post as spam/abuse.


  • 0 in reply to Mauro Kerry
    TI__Guru* 91086 points

    check out slide 3 and on  - this is what you see?

    8080.Notes on TMS3705A1DRG4 & Read.pdf

  • 0 in reply to Josh Wyatt
    Expert 1130 points
    Now I’m really confused; if what Kerry says is true ... why have not any-one else seen this undocumented behavior?

    So Kerry, what did you do to make HDX work.

    Would you please temporary remove the original experimenter coil, and replace it with a Resistor of say 1Kohm. Do the “extra” oscillations still “linger on” for 1mS after TxCT returns to its high state? (Into a non-resonant load).

    Josh, is this behavior mentioned anywhere in the TMS3705 data sheet; can you also do some tests to?

    Kerry, how can I get to “see” your waveforms ... I can’t seem to find them anywhere?

    Meantime, these are my new thoughts; regardless that you have found a working fix: Sorry is bit convoluted.

    Kerry now seems determined that Ant1 and Ant2 keeps oscillating “long after (1mS)” TxCT returns to a high state, and Josh seems to suggest he is actually looking at abnormal classic exponential decay. I can see a case for decaying oscillatory coil current but not voltage across Ant1 and Ant2.

    This exponential decaying current frequency will almost certainly be well different from the crystal derived 134.2KHZ Bridge drive inputs. (and be environment dependent?)

    If at the end of TxCT Ant1 and Ant2 both revert to ground state, how can this situation then be? Kerry, please confirm VDAA and VSSA are correct, and C4 is a low ESR capacitor for high pulse currents.

    The cleverness of the TMS3705 is that it is ONLY the indeterminate free-running resonance frequency of the approaching transponder coil (AND NOTHING ELSE) ... that should determine the signature definition of what represents a Binary One and Binary Zero.

    So Josh, are you thinking the “snapshot frequency capture” taken at 1.9mS (after the rising edge TcXT), should only see nothing (if no transponder present) OR a stream of “monotonous “ stable incoming pre-bits (arriving from a single approaching the Transponder) ... and note well, I suggest only consider a SINGLE transponder telegram case.

    The pivotal cleverness of the TMS3705 is that long before the arrival of the Transponders telegram’s “start byte”, its internal algorithm(s) has ALREADY learnt and stored what defines the structure of ZERO Bit and what represents structure of a ONE Bit. (... and a tolerance band to accept as representing a ONE ... clever ... +5 or +7 ... more later?). The internal algorithm will do it all again for the next single tag that comes along?

    Josh is it possible here, that the stored definitions of a ZERO Bit and a ONE Bit are in error? Being NOT derived from the now arriving stable Transponder pre-bits, BUT derived from the unpredictable/overpowering ringing of the Air-Gap-Resonant Transformer: These cross referring/correlation/masks (for bit pattern matching pattern) ... are totally wrong.

    This telegram decoding confusion is happening because in the 1.9mS window/aperture waveform is too complex, being a vector mixture of at least three unsynchronised resonances, or is totally dominated by the wrong feature?
    My metaphor here, is you have a Red filter on your right eye and a Blue filter on your left eye and now the words on the book you are reading are coloured Pink and Green? (err, sorry...) you read something but it’s all garbled!

    The Transponder is a simple “pinging” free running L/C oscillator and subject to frequency detuning with component tolerances AND surrounding sea water impedance ... and the detuning ferrite mineralization in bottom of the Stream/estuary? ... (Little low flying fish?)

    Even if, by beyond rare chance, the Main Resonances were both 134.2KHz, it is even less likely the 123KHZ would also be identical... see tuning capacitor tolerance(s).

    In an ideal “single coil reader” , Long before the arrival of the “START BYTE” and long after the “END OF THE CHARGE PHASE”, that is within a nominal 3mS “guard band”... more precisely, starting 1.9mS after the charge phase ends... the TMS has learnt how to recognize a FSK depiction of a ONE or ZERO. (FOR ONLY THIS CURRENT TAG ... FOR THE NEXT TAG A NEW DEFINITION OF ONE AN ZERO WILL BE COMPOSED)

    Kerry, have you now resonated the separate TMS3705 Receiver coil to 129KHz AND Critically Damped the separate Big loop’s High power Transmitter Transformer? (and left it resonant to 134.2 KHz?).

    See the RFM007 where (at the end of the charge phase) they literally MOSFET switch a high wattage damping resistance across the only coil AND simultaneously MOSFET switch a re-tuning capacitor to change the SAME coils resonance to 129KHz.

    I will review things now as I wait for your comment. My motive here is I would like to know your understanding of what will happen if two or more Transponders enter the detection field in “lock step”? I am still not confident enough with TMS3705, but do want to eventually use it. (as does Kerry).

    Kind Regards

    Ray
  • 0 in reply to Ray Seidel
    TI__Guru* 91086 points

    Ray - 

    the pre-bit portion of the response from the tag is 16 bits of "low" signals before the start byte. this is using low bit frequency (=134,2kHz), so, if each bit has 16RclkL cycles then (1/134.2kHz * 16) * 16 = 1.9mSec. this is documented in RFM Sequence Control manual and other spots too i think, like all the tag data sheets. Hope that clears that up. 

    if you look at Figure 6 (page 16) of the linked doc, it shows the waiting time that results, as all one really cares about is that start byte to know you have tag data incoming. 

    the question about the collisions i think answers itself, this is the reason HDX+ exists (to do anti-collision)

    hope that helps out. 

    BTW, the TRF4140 is a released device and is capable of doing both HDX and FDX - its not on ti.com, and currently needs some NDA paperwork in place to get it - but send me email and that process can get started if you are interested.

    if not, TMS3705 is the only other choice really since the RI-RFM-006 is not available anymore. (it was also capable of doing both frequencies used for LF)

     

  • 0
    Prodigy 115 points
    HI Josh,

    I would definitely be interested in the TRF4140.  So, let me know how I can get info, etc. on it.  Would be nice to have one part for both formats.

    Kerry

     



    On Thursday, March 26, 2015 4:47 AM, Josh Wyatt <bounce-614@mail.e2e.ti.com> wrote:


     
    A Message from the TI E2E™ Community
    Texas Instruments
     
    		 Josh Wyatt
    Ray - 
    the pre-bit portion of the response from the tag is 16 bits of "low" signals before the start byte. this is using low bit frequency (=134,2kHz), so, if each bit has 16RclkL cycles then (1/134.2kHz * 16) * 16 = 1.9mSec. this is documented in RFM Sequence Control manual and other spots too i think, like all the tag data sheets. Hope that clears that up. 
    if you look at Figure 6 (page 16) of the linked doc, it shows the waiting time that results, as all one really cares about is that start byte to know you have tag data incoming. 
    the question about the collisions i think answers itself, this is the reason HDX+ exists (to do anti-collision)
    hope that helps out. 
    BTW, the TRF4140 is a released device and is capable of doing both HDX and FDX - its not on ti.com, and currently needs some NDA paperwork in place to get it - but send me email and that process can get started if you are interested.
    if not, TMS3705 is the only other choice really since the RI-RFM-006 is not available anymore. (it was also capable of doing both frequencies used for LF)
     
    Did this answer your question?
     
    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.

    Flag this post as spam/abuse.


  • 0 in reply to Josh Wyatt
    Expert 1130 points
    Kerry, your observation is "probably correct" (about cycle over-runs of Ant 1-2?)
    I found scba031.pdf ... see para 3.22 and 3.23
    Fig 9 Shows some significant "internal latency of TxCT.
    Guess you cannot/should-not rely on using time measurements "from rising edge of TxCT ?
    Using UCC27424 power booster reduces this latency to 15uS.
    All just interesting ,if trying to do so something just bit different .
    Kind Regards Ray
  • 0 in reply to Josh Wyatt
    Prodigy 140 points
    Hi Josh

    I am also interested in the TRF4140, please send me more info on the chip and how to obtain it
  • 0
    Prodigy 115 points
    Same here.

    Kerry


    On Monday, February 1, 2016 10:34 AM, Christo Wiese <bounce-4596984@mail.e2e.ti.com> wrote:


     
    A Message from the TI E2E™ Community
    Texas Instruments
     
    Hi Josh

    I am also interested in the TRF4140, please send me more info on the chip and how to obtain it
     
     
    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.
    Flag this post as spam/abuse.