|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
I think most of our problems with our customised Silvertone Transmitter were due to my not being very RF-Literate. The quality of the equipment is great, and after many discussions with Bob from Silvertone, I'm convinced he knows his stuff when it comes to radio circuit design.. he sells a lot of gear to the military and commercial sectors, despite it being fairly pricey compared to the japanese gear, so there must be something going for it.
We had the transmitter mounted in a non-standard case with some odd grounding issues that initially messed up the radiated power. After we realised that and tuned the antenna to match, things improved a lot, but still werent working very well.. But then again, neither were my standard 36mhz JR radios, so it wasnt just the Silvertone's fault.
I think some of the problems were due to the fact that the Silvertone was running on 29Mhz (the "legal" ground frequency in Australia)
From what we've discovered so far, the best economical option appears to be the 3 Channel 75Mhz AM Radios from Tower Hobbies in the US for $60 (US) each. They're about $100 by the time they're landed here.
The 75mhz frequency allows antennas that are half the length of the 29/36Mhz antennas to receive just as good a signal as the longer ones, plus it allows use of the even shorter "deans" base-loaded antennas that all the americans use with great results, and the higher frequency seems to help make them less succeptible to interference from the motors.
We also suspect that the AM receivers, contrary to popular wisdom, may actually be better than the FM ones for our application. At least, our experience seems to show that, and despite everyone thinking they "know that FM is better than AM (do some research into it and you'll find out why it isnt true), we have a theory to do with the AM's AutoGainControl (AGC) circuit at short ranges that may explain why.
Anyway, back to the point, If you have to go with 29/36 mhz gear, and are considering the Silvertone, I can say that its quality gear and I expect that it would work as well or better than any of the imported gear, despite our experienced problems (which we had with ALL the low frequency FM radios) _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Thu Oct 07, 2004 11:10 am |
|
|
|
|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
Yes, its because most peoples experience with "radio" is limited to their little music box with either 3KZ or MMM FM. and for that application, the wide-band signal used in "FM" does sound heaps better than the narrow-band "AM", so naturally they assume FM is better than AM.
In *some* circumstances FM *is* better, all other things being equal, but in the case of hobby radio's all other things are *not* equal.
First of all, the "FM" implementation in FM hobby radio's is actually a narrow-band FM, with modulation of only a few Kilohertz (compared to broadcast radio which is something like 100Khz IIRC). Its proper name is NBFSK, for Narrow-Band-Frequency-Shift-Keying.
Of course thats too much of a mouthful for the general public, and seeing as how they all knew that FM was better, the hobby industry decided to call the new modulation scheme "FM" to take advantage of an already known term that everyone liked. Much easier to convice them to throw their old radios away and get the "cleaner signal" of FM.. after all, doesnt your FM music radio sound better ?? Typical Marketing Department tactics.
As far as range goes, I take it you are using Servo Switches in your bot ? Its easy to get good range with them, since they have no failsafe function, so any noisy signals are just ignored by the servo's. Unfortunately, they're not legal to compete with, so you need an ESC with failsafes, and the failsafes work by looking for invalid servo signals and shutting the ESC down when they are received. The problem with this is, that as soon as the signal gets noisy, the ESC starts jittering, making the bot much more succeptible to radio interference than a servo switched bot. _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Thu Oct 07, 2004 6:24 pm |
|
|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
Reality often doesnt agree with what should happen..
and remember that "FM" Hobby radios are NOT "FM" in the sense of a normal variable modulation FM signal. They are Frequency Shift Keyed. Which to non techie means they listen for the carrier frequency to move up or down by a certain amount to decide if the output is going to be high or low.
Our present theory is that at the extremely short ranges we typically operate at, the receiver has plenty of signal, but its accompanied by a *lot* of noise from the motors.
Apparently FM receivers run at full receiver gain all the time, which means all this noise gets amplified as well and sent to the frequency-shift-keying decoder, which swamps it with a lot of wide-band noise, possibly causing it to register some of this noise as a FSK bit, when in actual fact the Tx is not outputting that at all. Result, garbage data, invalid servo timing and a failsafe trigger.
The "AM" receivers have an Auto-Gain-Control (AGC) circuit that adjusts the sensitivity of the receiver according to how strong a signal it is seeing. At the ridiculously short ranges we operate at (<15 feet), the Rx has plenty of signal, so it winds down the amplification of the receiver, which also pulls down the amplification of the background noise. Operating at this lower gain level, the receivers are apparently less sensitive to local noise sources.
Thats our theory anyway. Until we can find someone with a spectrum analyzer who is willing to spend some time looking into whats happening in our receivers, we're reduced to just using what seems to work and trying to figure out why it does. _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Thu Oct 07, 2004 9:26 pm |
|
|
|
|
|
|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
nope, cause all you're doing then is making the signal/noise ratio worse. weakening the signal, but the noise stays the same and is still amp'ed by the Rx.
You need to improve the "selectivity" of the receiver I think.. do a google for "radio selectivity" and youll find lots about it
I recall a job a few years back with some radio alarm gear working from boats out on the bay, back to a monitoring station on the pier. was suffering intereference from nearby pager transmitters.. we put better antenna's on, twiddled with the Tx power and generally bumbled about ignorantly, until we gave up and called in a radio tech, who took a few measurements and promptly put an *attenuator* in the receivers antenna line. Problem solved.
Apparently by increasing the Rx gain with better antennas and whatnot, we were just deafening the receiver with the increased background noise as well.. reducing the signal allowed the limited "selectivity" decoder to work better.. baffled the hell out of us.. since when do you make a radio work better by *weakening* the signal ? I think I decided I'd stick to digital logic there, at least it follows sensible rules.. _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Fri Oct 08, 2004 1:40 am |
|
|
|
|