Post by thread:[OT]: Building a quick-and-dirty wireless modem (w

Hi List, About ten years ago I built a simple super-regenerative FM receiver, which worked remarkably well, considering that I used a single transistor and maybe eight passives. I was wondering if I can build a quick-and-dirty wireless modem based on a super-reg, a one-transistor AM transmitter, and a modem IC. I have a project that requires a speed of only 1200 baud, half duplex, and a range of 10-15 meters. Frequency stability is a priority. I was looking at the TCM 3105 (which is relatively expensive despite its age, btw), and remembered how Russel suggested a while back that one could use a DTMF circuit instead of a "real" modem IC. What are your thoughts on that? What is the highest baud rate that one can theoretically achieve with a DTMF chip? Or maybe there are less expensive real modem ICs out there? Best regards, Vitaliy Alison Lewis wrote: {Quote hidden}> If you've been watching, there were some great RF > devices suggested to me, so I put the list together > for anyone else who may be looking for small RF > transmitters. I went with > http://www.rentron.com/remote_control/TXLC-315.htm. > > > http://www.sparkfun.com > www.radiotronix.com/products/prodrctas.asp > http://www.rentron.com/remote_control/TXLC-315.htm or > www.rentron.com/remote_control/TXLC-434.htm > http://www.rfsolutions.co.uk

On November 25, 2005 04:29 pm, Vitaliy wrote: > Hi List, > > About ten years ago I built a simple super-regenerative FM receiver, > which worked remarkably well, considering that I used a single > transistor and maybe eight passives. I was wondering if I can build a > quick-and-dirty wireless modem based on a super-reg, a one-transistor > AM transmitter, and a modem IC. Sure you can. You just finished saying you built a receiver 10 years ago. > I have a project that requires a speed of only 1200 baud, half > duplex, and a range of 10-15 meters. Frequency stability is a > priority. Sounded okay until you said frequency stability is a priority, which frequency needs to be stable? the regen or the 1200baud? If you meant the regenerative receiver, no, you can't because by nature a regen receiver locks to the strongest transmission, which is not necessarily your own transmitter. If you meant the 1200baud, then that is dictated by the transmitter unless your receiver is made to create it's own 1200baud for "no signal". > I was looking at the TCM 3105 (which is relatively expensive despite > its age, btw), and remembered how Russel suggested a while back that > one could use a DTMF circuit instead of a "real" modem IC. What are > your thoughts on that? What is the highest baud rate that one can > theoretically achieve with a DTMF chip? Have your TX circuit send AM as you planned, and have your receiver receive AM and send the AM into a comparator (some PIC chips have comparators built in, for example pic16c620 ....but choose a more updated chip since the 620 is no longer built). Then have your pic chip generate the 1200baud. > Or maybe there are less expensive real modem ICs out there? Probably, but I didn't look at any of the suggestions mentioned below. {Quote hidden}> Alison Lewis wrote: > > If you've been watching, there were some great RF > > devices suggested to me, so I put the list together > > for anyone else who may be looking for small RF > > transmitters. I went with > > www.rentron.com/remote_control/TXLC-315.htm. > > > > > > http://www.sparkfun.com > > www.radiotronix.com/products/prodrctas.asp > > http://www.rentron.com/remote_control/TXLC-315.htm or > > www.rentron.com/remote_control/TXLC-434.htm > > http://www.rfsolutions.co.uk

Jose Da Silva wrote: >> transistor and maybe eight passives. I was wondering if I can build a >> quick-and-dirty wireless modem based on a super-reg, a one-transistor >> AM transmitter, and a modem IC. > > Sure you can. > You just finished saying you built a receiver 10 years ago. Good one. :-D If you weren't trying to be funny, I think there's more to a wireless modem than a superregenerative receiver. >> I have a project that requires a speed of only 1200 baud, half >> duplex, and a range of 10-15 meters. Frequency stability is a >> priority. > > Sounded okay until you said frequency stability is a priority, which > frequency needs to be stable? the regen or the 1200baud? Regen. > If you meant the regenerative receiver, no, you can't because by nature > a regen receiver locks to the strongest transmission, which is not > necessarily your own transmitter. Since we're talking about a 10-15 meter range, I can probably ensure that my transmitter is the loudest signal in the room. > Have your TX circuit send AM as you planned, and have your receiver > receive AM and send the AM into a comparator (some PIC chips have > comparators built in, for example pic16c620 ....but choose a more > updated chip since the 620 is no longer built). > Then have your pic chip generate the 1200baud. Hold on.. do you mean 1 = signal, 0 = no signal? That's not what I had in mind, I was thinking in terms of FSK or PSK amplitude-modulating the carrier. The simplified requirements are: I need two black boxes that would let me connect two terminals 10 meters apart, and allow communication at 1200 baud. I know you guys tend to assume there's a pic on each end that I can squeeze some modem code into, but that's not the case here. :-) I would prefer a "transparent" hardware solution, but if there is open source modem code for a low-end PIC, I would consider that as well. >> Or maybe there are less expensive real modem ICs out there? > > Probably, but I didn't look at any of the suggestions mentioned below. Links below are for transmitters/receivers and wireless modems, not modem ICs. Best regards, Vitaliy {Quote hidden}>> Alison Lewis wrote: >> > If you've been watching, there were some great RF >> > devices suggested to me, so I put the list together >> > for anyone else who may be looking for small RF >> > transmitters. I went with >> > www.rentron.com/remote_control/TXLC-315.htm. >> > >> > >> > http://www.sparkfun.com >> > www.radiotronix.com/products/prodrctas.asp >> > http://www.rentron.com/remote_control/TXLC-315.htm or >> > www.rentron.com/remote_control/TXLC-434.htm >> > http://www.rfsolutions.co.uk > --

On November 25, 2005 07:07 pm, Vitaliy wrote: > Jose Da Silva wrote: > >> transistor and maybe eight passives. I was wondering if I can > >> build a quick-and-dirty wireless modem based on a super-reg, a > >> one-transistor AM transmitter, and a modem IC. > > > > Sure you can. > > You just finished saying you built a receiver 10 years ago. > > Good one. :-D > > If you weren't trying to be funny, I think there's more to a wireless > modem than a superregenerative receiver. For a digital circuit, yes, there is slightly more. An AGC is needed unless you do FM (my preference would be to send 2-tones-FSK and then use the PIC to differentiate the 2-tones, this way the AGC is useful but isn't as critical). {Quote hidden}> >> I have a project that requires a speed of only 1200 baud, half > >> duplex, and a range of 10-15 meters. Frequency stability is a > >> priority. > > > > Sounded okay until you said frequency stability is a priority, > > which frequency needs to be stable? the regen or the 1200baud? > > Regen. > > > If you meant the regenerative receiver, no, you can't because by > > nature a regen receiver locks to the strongest transmission, which > > is not necessarily your own transmitter. > > Since we're talking about a 10-15 meter range, I can probably ensure > that my transmitter is the loudest signal in the room. Radio passes through walls. Your 100mW may still be overshadowed by a transmitter a mile away pumping out 5W.... unless your frequency of choice is dedicated to low-power stuff, for example garage-door openers or particular bands like that. If you use those bands, then, yes, you're likely okay. I'm not sure why you need freq stability. > > Have your TX circuit send AM as you planned, and have your receiver > > receive AM and send the AM into a comparator (some PIC chips have > > comparators built in, for example pic16c620 ....but choose a more > > updated chip since the 620 is no longer built). > > Then have your pic chip generate the 1200baud. > > Hold on.. do you mean 1 = signal, 0 = no signal? That's not what I > had in mind, I was thinking in terms of FSK or PSK > amplitude-modulating the carrier. AM is what I thought you meant from your original posting. FM FSK PSK is much much better, so I'm glad you chose it. > The simplified requirements are: I need two black boxes that would > let me connect two terminals 10 meters apart, and allow communication > at 1200 baud. I know you guys tend to assume there's a pic on each > end that I can squeeze some modem code into, but that's not the case > here. :-) Well, it is the piclist, so it was a guess a pic was needed here, but something in your circuit needs to deal with radio noise, be it the black-box, the pic, or the terminal at the end of the 1200baud.

> For a digital circuit, yes, there is slightly more. > An AGC is needed unless you do FM (my preference would be to send > 2-tones-FSK and then use the PIC to differentiate the 2-tones, this way > the AGC is useful but isn't as critical). AGC is useful, and it's cheap -- so I am planning to use it. > Radio passes through walls. Yes, I noticed. ;) > Your 100mW may still be overshadowed by a transmitter a mile away > pumping out 5W.... Wow, you are right! I should have done my homework... this goes against my intuition: 100 mW / (10 m)^2 = 1 mW/m^2 5000 mW / (1609 m)^2 = 1.9 mW/m^2 > unless your frequency of choice is dedicated to > low-power stuff, for example garage-door openers or particular bands > like that. If you use those bands, then, yes, you're likely okay. Alright. Where would I find a list of such frequencies, and how would I know what would make it legal to transmit at those frequencies (output power, antenna length)? > I'm not sure why you need freq stability. I haven't done anything wireless for a long time, but I do remember how I had to constantly adjust the trim cap (or stretch/compress the coil) to compensate for frequency shift due to differences in temperature. That's what I meant by frequency stability -- not having to tune the Tx/Rx modules every time I need to use them. > AM is what I thought you meant from your original posting. > FM FSK PSK is much much better, so I'm glad you chose it. Yes, I meant "AM modulated carrier." > Well, it is the piclist, so it was a guess a pic was needed here, but > something in your circuit needs to deal with radio noise, be it the > black-box, the pic, or the terminal at the end of the 1200baud. I was surprised to find that there isn't much choice when it comes to 1200 baud modems. TCM3105 seems to be the easiest one to use, unfortunately the datasheet says "Full Duplex up to 1200 Tx and 150 Rx" I was planning to use it with a circuit which automatically echoes back whatever it receives, so I need FD at 1200 baud both ways. It seems that my only choice at this point is to use XR2206 and XR2211. --- took a break to do more Googling --- What if I use an rfPIC, and a stand-alone RF receiver (e.g., rfRXD0420)? Vitaliy

On Sun, 27 Nov 2005 09:56:28 -0200, Gerhard Fiedler wrote: > Vitaliy wrote: > > >> Your 100mW may still be overshadowed by a transmitter a mile away > >> pumping out 5W.... > > > > Wow, you are right! I should have done my homework... this goes against my > > intuition: > > > > 100 mW / (10 m)^2 = 1 mW/m^2 > > 5000 mW / (1609 m)^2 = 1.9 mW/m^2 > > Maybe your intuition wasn't that wrong, after all... Probably when you > divided 5000 by 2588881 (that's 1609^2), you took the 10^-3 of the result > as meaning "milli", but with milli already going in, it's "micro": > > 5000 mW / (1609 m)^2 = 1.9 uW/m^2 Hang on, chaps! While the ratio of field strengths in the above calculation is correct (1 : 0.0019) it's wrong to give them a dimension - the actual field strengths certainly won't be the figures shown above. If you assume an antenna with no gain (isotropic, giving a perfectly spherical radiation pattern) field strength of 100mW at 10m will be about 0.079 mW/m^2 (calculated using the area of a sphere, 4 pi r^2). Now isotropic antennae are kept in the same drawer as frictionless pulleys and light, inextensible string - they are theoretical only, but a practical antenna for the appication under discussion is unlikely to have a gain over isotropic of more than about 2 or 3. You'd need a highly directional antenna with a gain of about 12 to get 1 mW/m^2, and it's very unlikely that that's what would be used in this case. Cheers, Howard Winter St.Albans, England

Gerhard Fiedler wrote: > Maybe your intuition wasn't that wrong, after all... Probably when you > divided 5000 by 2588881 (that's 1609^2), you took the 10^-3 of the result > as meaning "milli", but with milli already going in, it's "micro": > > 5000 mW / (1609 m)^2 = 1.9 uW/m^2 This is the second time my math skills embarrass me.. and both times it was you who pointed out the mistake. ;-D Then Howard Winter wrote: {Quote hidden}> Hang on, chaps! While the ratio of field strengths in the above > calculation is correct (1 : 0.0019) it's > wrong to give them a dimension - the actual field strengths certainly > won't be the figures shown above. > > If you assume an antenna with no gain (isotropic, giving a perfectly > spherical radiation pattern) field > strength of 100mW at 10m will be about 0.079 mW/m^2 (calculated using the > area of a sphere, 4 pi r^2). Now > isotropic antennae are kept in the same drawer as frictionless pulleys and > light, inextensible string - they > are theoretical only, but a practical antenna for the appication under > discussion is unlikely to have a gain > over isotropic of more than about 2 or 3. You'd need a highly directional > antenna with a gain of about 12 to > get 1 mW/m^2, and it's very unlikely that that's what would be used in > this case. What you are saying about field strengths makes perfect sense, but I guess in this particular case I was only interested in the ratio. Jose Da Silva said: > Your 100mW may still be overshadowed by a transmitter a mile away > pumping out 5W.... Which is obviously wrong, assuming the same radiation pattern for both antennas a transmitter one mile away would have to output 2.6 kW to match the signal of the local 0.1 W transmitter (Gerhard, let me know if my calculations are wrong again ;) Vitaliy

Vitaliy wrote: > Gerhard Fiedler wrote: >> 5000 mW / (1609 m)^2 = 1.9 uW/m^2 > > [...] and both times it was you who pointed out the mistake. ;-D I wish I could have contributed something more substantial :) {Quote hidden}> What you are saying about field strengths makes perfect sense, but I > guess in this particular case I was only interested in the ratio. Jose > Da Silva said: > >> Your 100mW may still be overshadowed by a transmitter a mile away >> pumping out 5W.... > > Which is obviously wrong, assuming the same radiation pattern for both > antennas a transmitter one mile away would have to output 2.6 kW to > match the signal of the local 0.1 W transmitter (Gerhard, let me know if > my calculations are wrong again ;) Not really :) Just adding that you could factor in that the "other" antenna is highly directional and yours is isotropic; then the 2.6 kW could go down by the ratio between their and your antenna. Howard mentioned that there are antennas with a gain of 12, and possibly it could be even higher. So a transmitter with 216 W or less could do it -- and, depending on the application, it also may not be a full mile away. Bringing that together it seems that in the end you could get in trouble with a 5 W transmitter, with a highly directional antenna, 250 m away... (assuming that same RF power means trouble, which I don't know whether it does). Gerhard

On November 27, 2005 02:09 am, Vitaliy wrote: > > Radio passes through walls. > > Yes, I noticed. ;) 8-) > > Your 100mW may still be overshadowed by a transmitter a mile away > > pumping out 5W.... > > Wow, you are right! I should have done my homework... this goes > against my intuition: > > 100 mW / (10 m)^2 = 1 mW/m^2 > 5000 mW / (1609 m)^2 = 1.9 mW/m^2 Friend of mine was trying out some industrial controls. He managed to still control the system from over 15 miles away. He had to talk with his partner over a cell phone to make sure he didn't accidently run the machinery somewhere unwanted, but the distance was the interesting point. 5W, 450something MHz. Not theory, but actual fact. Your calculations are valid for a point antenna but disregards the fact that directional antennas exist and may or might not cross your property or the site of interest. ;-) Like Gerhard points out, a directional antenna crossing the property would cause problems.... but reading your reply to Olin under EE, sounds like this is a project under the fun-factor, so no point in butting horns, have fun. ;-) > > unless your frequency of choice is dedicated to > > low-power stuff, for example garage-door openers or particular > > bands like that. If you use those bands, then, yes, you're likely > > okay. > > Alright. Where would I find a list of such frequencies, and how would > I know what would make it legal to transmit at those frequencies > (output power, antenna length)? http://www.google.ca/search?hl=en&ie=ISO-8859-1&q=radio+frequency+chart+pdf&btnG=Google+Search&meta= ...some countries tend to follow the US but not necessarily. fcc tends to look the other way for 100mW or less, unless you cause problems. > > I'm not sure why you need freq stability. > > I haven't done anything wireless for a long time, but I do remember > how I had to constantly adjust the trim cap (or stretch/compress the > coil) to compensate for frequency shift due to differences in > temperature. That's what I meant by frequency stability -- not having > to tune the Tx/Rx modules every time I need to use them. Makes sense, thanks. One place I used to work used thermistors to aid for compensation, but that tended to be a custom solution. If this is a one-of-a-kind, you could probably go that way, but if you are building many, you might go the way of using a varactor and a temperature chart if building many. Apply more/less voltage to a varactor to tune your LC. PIC holds the chart and could bleed in the varactor voltage thru a 1meg resistor. > > AM is what I thought you meant from your original posting. > > FM FSK PSK is much much better, so I'm glad you chose it. > > Yes, I meant "AM modulated carrier." FSK would be preferable if you have to deal with long strings of 1s or 0s, this way you tend to have 50%high/50%low. AM tends to drift to either high or low (depending on how you tuned your circuit). For example, say your circuit is usually 5v for nothing sent, then when you send a long string of 0s, you'll note your signal drifts back towards 5v despite fact you continue to send 0s. I noticed in Olin's reply (tag changed to EE he speaks of manchester encoding, but it's still a form of 50%high/50%low which is what you would want to do, but as this is a project under the fun-factor, choose what you like. For what you want to do, you may find it simple to send FSK, maybe 2400hz for a 0 and 4800hz for a 1. Watch for 0->1 or 1->0 crossings. > > Well, it is the piclist, so it was a guess a pic was needed here, > > but something in your circuit needs to deal with radio noise, be it > > the black-box, the pic, or the terminal at the end of the 1200baud. > > I was surprised to find that there isn't much choice when it comes to > 1200 baud modems. TCM3105 seems to be the easiest one to use, > unfortunately the datasheet says "Full Duplex up to 1200 Tx and 150 > Rx" I sort of expected that without bothering to look. Something this "fine" you are unlikely to find as open source ideas and you are more likely going to have to roll your own solution or ask someone to code a black box for you. > I was planning to use it with a circuit which automatically echoes > back whatever it receives, so I need FD at 1200 baud both ways. It > seems that my only choice at this point is to use XR2206 and XR2211. > > --- took a break to do more Googling --- > > What if I use an rfPIC, and a stand-alone RF receiver (e.g., > rfRXD0420)? rfpic has got some good open source ideas in the pdfs you can follow. Looks like approx $3 in parts starting from your regen output (or standalone receiver) going to your rfpic 1200 baud output. The rfpic should be able to clean up the signal for you and present a nice 1200baud solution for you to use, assuming the transmitter sends a decent packet of information (555555-data-checksum). If you have your own AGC and it's cheap, you might try a seperate pic and your AGC circuit, that may bring you down to $1 for the pic and whatever your AGC circuit costs, maybe less than $2 altogether. In standalone mode, you might want the rfpic to also send a carrier detect or RTS, or maybe better, a DTR, to your terminal. DTR may be better, since if your receiving pic can't receive a signal, then there is a good chance you can't transmit the other way either until the radio path is okay for transmission again. ...just food for thought, and have fun with your project.