Wavess

Another Electro Telemaster FPV

85 posts in this topic

What model number is the servo? Did it directly fit in the Pandora Pan & Tilt mount without any changes?

Thanks

My model is Hyperion DS13-TMB but I think that you could go with a range from the 12 to 13 series, they are all same lenght. You´d need USB cable for programming.

Share this post


Link to post
Share on other sites

Thanks, I'll look into the Hyperions.

Share this post


Link to post
Share on other sites

Reviewing yesterday video seems that I have 2 problems.

One is the pan servo is constantly vibrating seems like there´s some kind of rf interference, didn´t notice yesterday flying without fpv.

Second, seems that tinymic is saturated by the wind, on its location, I´ll adjust the level pot and see if I could hide inside my wind.

There´s also a little effect with white horizontal lines very thin on the video, I suposse this is due to cable lenght. I´d install some caps close to the tx to see if helps.

Surprises me to have glitches with this components dispossal. I feel disarmed with the rf noise :o

Share this post


Link to post
Share on other sites
One is the pan servo is constantly vibrating seems like there´s some kind of rf interference, didn't notice yesterday flying without fpv.

I've noticed that many servos are glitchy from the Tx's RF. The Hitec HS81 and HS65 are decently immune, which is why I like them. The first thing to try with your Hyperions would be some toriod cores with several tight wraps of servo cable on them.

seems that tinymic is saturated by the wind, on its location, I´ll adjust the level pot and see if I could hide inside my wind.

Tiny-Mic's volume control should save the day. However, adding a little bit of foam to it's face can also aid in reducing the wind noise.

There´s also a little effect with white horizontal lines very thin on the video, I suppose this is due to cable length.

Many A/V Tx's will do that when the Audio level is too high or if the Tx is exposed to vibration. If the noisy lines increase as the throttle is advanced, and go away with the motor off, then those two issues are probably biting you. If the noise occurs at all times, then I would suspect that a switching Vreg is powering the A/V gear. Is one being used?

Share this post


Link to post
Share on other sites
I've noticed that many servos are glitchy from the Tx's RF. The Hitec HS81 and HS65 are decently immune, which is why I like them. The first thing to try with your Hyperions would be some toriod cores with several tight wraps of servo cable on them.

I don´t know if there is only a servo problem, I also heard a motor glitch once. Would be difficult to add now toroid cores on my cable since all of them is twisted one, but I´ll see how to do it, do u think it´s better close to rx or close to servo?

Many A/V Tx's will do that when the Audio level is too high or if the Tx is exposed to vibration. If the noisy lines increase as the throttle is advanced, and go away with the motor off, then those two issues are probably biting you. If the noise occurs at all times, then I would suspect that a switching Vreg is powering the A/V gear. Is one being used?
Has to be something related with my long wires, I had the same tx on my easystar without a single problem. It´s not an audio problem because the lines are always there (you can almost notice them, but they are) and I´m using linear reg on it. BTW the lin reg is close to my tx, on the same pcb.

Share this post


Link to post
Share on other sites
I don´t know if there is only a servo problem, I also heard a motor glitch once.

It might be wise to revisit the Switching BEC on the R/C Rx.

do u think it´s better close to rx or close to servo?

It depends. In your situation, I would try them on both ends. BTW, if the cables are too short or already twisted, then a quick test is to grab some extensions and wrap them through the toriod with many turns. Then temporarily try them on the model. If the problem goes away, you will know what to do. But in your case, I think you have more evilness to fix.

I´m using linear reg on it. BTW the lin reg is close to my tx, on the same pcb.

Does the Linear VReg have an adequate stability cap on its input? Long wires to a VReg usually require that. Have a look at the VReg's data sheet and see what is suggested. Plus, add toroids to the VReg's input and output too.

Share this post


Link to post
Share on other sites
It might be wise to revisit the Switching BEC on the R/C Rx.

Yes I think that the best way here it´s to install a separate 4 AA battery and disable the ubec for a flight and see what happens

It depends. In your situation, I would try them on both ends. BTW, if the cables are too short or already twisted, then a quick test is to grab some extensions and wrap them through the toriod with many turns. Then temporarily try them on the model. If the problem goes away, you will know what to do. But in your case, I think you have more evilness to fix.

That´s something I was thinking on doing, nice idea. But now one question that could open a new debate, theorically we have to find the resonant frecuency of the device to install the toroid´s proper value, or inductance. So probably a silly question, but would a frecuency counter do this job?

Does the Linear VReg have an adequate stability cap on its input? Long wires to a VReg usually require that. Have a look at the VReg's data sheet and see what is suggested. Plus, add toroids to the VReg's input and output too.

The linear reg, is mounted following the manufacturer data sheet, have a lot of caps every where, for the vreg and for the AWM 633

Forgot to answer you before, yes I installed hyperion without any mod to Pandora mount, but I´m only using 1 for pan, the tilt servos is a hitec hs-85mg. A member of our forum is using this hyperion servo without a problem.

Edited by Wavess

Share this post


Link to post
Share on other sites
theoretically we have to find the resonant frequency of the device to install the toroid´s proper value, or inductance. So probably a silly question, but would a frequency counter do this job?

The goal is to reduce the common mode EMI/RFI noise. It is usually broadband with microvolt amplitude. So if you want to "see" it you'll need a spectrum analyzer. Honestly, if it was me, I would just get as many wraps on the ferrite cores as possible (half dozen or more) and go from there. If you have an assortment of toroid cores, try the type 43 and type 77 materials.

Share this post


Link to post
Share on other sites

Well today I´ve tried more things on the backyard only

Removed de UBEC and aplied 4AA battery to the receiver. There is no difference in glitches, at least here.

Unplugged the video stuff to see if glitches become better, no difference

Tried with ferrite toroids, one kind is the one the LHS sells, and the other one is this:

http://es.rs-online.com/web/search/searchB...&R=467-4188

Was put on the +5V line to receiver, close to receiver, close to the batt, and on both sides, results, none.

Was put on the pan channel, close to the receiver, results none.

I´ve also observed that all channels suffer glitches, but the pan servo is the worst (maybe cause has more travel and it´s more noticeable.)

All this are not very encouraging :(

Share this post


Link to post
Share on other sites
Removed the UBEC and applied 4AA battery to the receiver. There is no difference in glitches, at least here. Unplugged the video stuff to see if glitches become better, no difference

Are there any other RFI/EMI sources in your installation? That is to say, what exactly is still plugged in after the other items have been removed? Have you tried another Rx? If the problem is isolated to the long servo leads, then there are different solutions to the issues that can cause trouble (due to cross-talk, voltage drop, EMI/RFI from the servo motor, etc.).

All this are not very encouraging

Solving EMI/RFI issues can be very frustrating because there is no single cure. The best solution is patience and methodical testing/experimenting. So, keep at it and take good notes as you go along.

BTW, did you find what is causing the unwanted lines in the video?

Share this post


Link to post
Share on other sites
Are there any other RFI/EMI sources in your installation? That is to say, what exactly is still plugged in after the other items have been removed? Have you tried another Rx? If the problem is isolated to the long servo leads, then there are different solutions to the issues that can cause trouble (due to cross-talk, voltage drop, EMI/RFI from the servo motor, etc.).

Unplugged is all the A/V related stuff, that is, camera, video TX, GPS, OSD.

Also unplugged the motor + ESC, without any difference.

I´ve not tried another Rx, don´t have more 8 channels rx on stock ;)

Would have to try with a 6 channels receiver at same conditions

Solving EMI/RFI issues can be very frustrating because there is no single cure. The best solution is patience and methodical testing/experimenting. So, keep at it and take good notes as you go along.

BTW, did you find what is causing the unwanted lines in the video?

No the video is not that bad, the lines appeared are very thin and almost can´t be seen. But I´d try to find the problem too when I got my r/c in order to make FPV. The only change I made on the video tx, when I moved from my easystar is that I had 2 linear DCDC, one step down to 5V and the other to 3.3v, all feeded by a 2S lipo. Right now I´ve skipped the 5v linear reg, and go with 2S straight to the 3.3 linear reg

Would continue to make more trials, and reporting them

Share this post


Link to post
Share on other sites

Well more trials today.

Tried unpluging all channels one by one to see if some servo is making the rx mad, but the behavior is similar.

Tried installing ferrite cores on all channels and seems to be a bit better, but not a real improvement.

Tried with a 6 channel ppm rx I got here, with the same conditions tried before with the original 8 channels, same stuff.

Tried also with a 8 channel receiver (same brand an model), same behavior.

And the last trial that has been the best, touching the rx antenna, all the glitches are gone away.

This makes me to conclude the following.

It´s an installation problem due to my long wires.

Seems to be a problem with the rx´s ground plane that´s fooled with my long wires, so ferrite cores will do nothing here, and perhaps forcing to have a dipole antenna, installing the same lenght on the antenna for the ground would be a posibility.

Share this post


Link to post
Share on other sites

I'm following along with this thread(as I often do. Who, me, lurk?) and I'm very curious about these problems you're running into. What frequency is your controller? 72mzh? 50mhz? I am thinking that an electro telemaster might just be my next FPV platform, so this thread particularly interests me. Good luck Wavess!

Share this post


Link to post
Share on other sites

35mhz here in Europe. Thanks a lot for your support

Edited by Wavess

Share this post


Link to post
Share on other sites

It's a shame you are still haunted by this problem. EMI/RFI can be a pain to solve, but there is always a solution.

Seems to be a problem with the rx´s ground plane that´s fooled with my long wires, so ferrite cores will do nothing here, and perhaps forcing to have a dipole antenna, installing the same length on the antenna for the ground would be a possibility.

It would be interesting to see what happens after you add the counterpoise element (poor man's dipole in your situation). Honestly, the glitches could easily become worse since you'll have more antenna to capture the localized EMI/RFI noise. But if it solves it, then that would be grand.

Since you've added toroids to the servo wires and the problem seemed to change a little, then that suggests that part of the problem is RFI coupled through, or radiating from, the servo wiring. A more aggressive solution would involve adding 10uH chokes on each of the servos' three wires, plus retain the toroids.

BTW, did you happen to get *at least* a half-dozen tight wraps (more is better) on the toriods, like the photo at the bottom of this page?: http://www.dpcav.com/xcart/product.php?pro...=255&page=1.

Most of your installation is a mystery to me, so it is hard to offer any particular advice. For example, R/C receiver's vary in their performance when they confront power supply and EMI/RFI noise. Also, the servos impact this too. For example, modern digital servos have very low impedance FET bridge amps and faster motor updates. This increases the noise floor in a R/C installation due to their higher instantaneous peak currents and faster rise times. That's not to say that legacy "analog" servos are free of this issue, but the point is that the new digital stuff did not help out. Long story short, the issue with glitches is affected by how and what you install.

If it was me, I would identify the most twitchy servo in the installation. Then, unplug all the other servos from the Rx and all other electronic (Tx, OSD, GPS, etc.) equipment. If the twitch problem goes away, then start plugging things back in, one at a time, too see the impact of each item. On the other hand, if the problem does not go away, then you have a perfect test platform to perform your noise reduction solutions on. That includes trying a different brands of Rx and servo, along with RF filters, servo isolators, servo buffers, and other assorted witchcraft.

If you cannot resolve the problem (that would be unusal), and the twitches are minor and do not affect flight, and your ground range is not affected, then you may want to consider a high quality PCM Rx. But, don't do that if the existing problem involves a substantial range loss with a good quality PPM Rx. You have to be very careful with such band-aides or you will get what you asked for, but not what you wanted. :)

Share this post


Link to post
Share on other sites
It would be interesting to see what happens after you add the counterpoise element (poor man's dipole in your situation).

:D I´m not a poor man, but not a rich either, :lol: just kidding. I think this solution would be one of the last I´d try, theorycally this has to be installated antenna to one side of the wing and ground plane on the other side, might be a bit hard to achieve it on telemaster

Since you've added toroids to the servo wires and the problem seemed to change a little, then that suggests that part of the problem is RFI coupled through, or radiating from, the servo wiring. A more aggressive solution would involve adding 10uH chokes on each of the servos' three wires, plus retain the toroids.

Do u think some cap would be needed too?

BTW, did you happen to get *at least* a half-dozen tight wraps (more is better) on the toriods, like the photo at the bottom of this page?: http://www.dpcav.com/xcart/product.php?pro...=255&page=1.

Yes, I post some pics at the end of this topic, you could see some of the ferrites I´m using (4 different kinds)

Most of your installation is a mystery to me, so it is hard to offer any particular advice. For example, R/C receiver's vary in their performance when they confront power supply and EMI/RFI noise. Also, the servos impact this too. For example, modern digital servos have very low impedance FET bridge amps and faster motor updates. This increases the noise floor in a R/C installation due to their higher instantaneous peak currents and faster rise times. That's not to say that legacy "analog" servos are free of this issue, but the point is that the new digital stuff did not help out. Long story short, the issue with glitches is affected by how and what you install.

But your expert advice is great to have.

I´ve noticed that the glicher servo is the digital pan (hyperion) but this could be understandable because it has 2 times the travel the other servos have.

If you cannot resolve the problem (that would be unusal), and the twitches are minor and do not affect flight, and your ground range is not affected, then you may want to consider a high quality PCM Rx. But, don't do that if the existing problem involves a substantial range loss with a good quality PPM Rx. You have to be very careful with such band-aides or you will get what you asked for, but not what you wanted.

Easy solution but that would be my last option, before I want to have a CLEAN plane in PPM, if I don´t solve this, I would think to replace components, BTW, I think this components placement is perfect, the only bad thing is the long servo wires. Right now range is acceptable, and the ground test I did at the flying field was good, I had control over plane more than what manufactures advice (more than 80 steps), but it´s very annoying to have the pan servo glitching.

One last thought, why do you think glitches are gone when I touch the antenna? because human body absorbs rf noise?

Here are the different ferrite cores I´m using + an inside pic of my installation. You´ll notice that I´m using also a plastic

tubular ferrite, this one I bought from LHS. On these pics I don´t have all ferrites installed, but when I tried I had.

post-1754-1206733510_thumb.jpg

post-1754-1206733547_thumb.jpg

Share this post


Link to post
Share on other sites
Do u think some cap would be needed too?

I don't believe so. If the problem was due to a peak current derived voltage brownout at the receiver, then a big cap there would help. But I don't think that is the case here. But, if you have a servo that is missing the little ceramic caps on its brushed motor, then that would explain some of the problem.

I´ve noticed that the glicher servo is the digital pan (hyperion) but this could be understandable because it has 2 times the travel the other servos have.

Grab the programmer and return it to normal travel. That will be more useful for your observations.

One last thought, why do you think glitches are gone when I touch the antenna? because human body absorbs rf noise?

Normally this sort of action would merely detune the antenna and make it less efficient (less sensitive). That is what I suspect is happening in your case, but it is hard to say.

Here are the different ferrite cores I´m using + an inside pic of my installation. You´ll notice that I´m using also a plastic tubular ferrite, this one I bought from LHS.

I don't see many wraps on the tubulars, as well as the donut toroid in the fuselage photo. Picture #2 was what I would use as a minimum wrap count for the donut types. You can get by with less wraps on a long tubular, but I'm mostly a donut user so that is where my main experience is.

With these things, more wraps means more common mode noise reduction. I usually do not see a significant impact until at least a half-dozen tight wraps are used. I've had instances were it took a lot more to do anything at all. Plus, the snap together split core types, as seen in photo 3, are about half as effective as one of similar size that is one piece. Lastly, the ferrite cores come in a variety of recipes; I tend to use type 43 and 77 for these applications. Of course the use of toroids, on their own, may not offer a remedy, but they won't hurt things at all.

Keep in mind that these sort of problems often do not have a single source that causes the problem. Sometimes its several related issues that, when taken together, cause a lot of headaches.

Share this post


Link to post
Share on other sites

After many and many testings on this issue, I´ve reached some conclutions on this problem:

Seemed that the 2 tail servos were causing some kind of interference due to their long cable and due to their almost parallel tracing with the 35mhz rx antenna wire. Well installed 2 toroids on both channels and now the situation don´t get worse when I connect them. But here a mistery, last day, I removed toroids, plug servos directly to rx, and seemed that everything were ok, so appeared that problem had gone without toroids, anyway since I had 1 issue the other day, now I remain using those toroids.

Found that my OSD, while I now have connected to my rx, (I didn´t have on previous Easystar´s config) there is some problems too, put 2 toroids, 1 on the power supply and the other one on the rx conection close to the rx, now seems that the problem has gone too, but depends on the day I test, sometimes I remain with some little difference between osd, on and off, and some days, there isn´t a difference, very weird.

With all the servos unplugged from the rx, there is a remaining glitch on pan servo, very little but there it´s. I installed other toroid on receiver´s end, and added 3 inductors close to the servo, tried with different values, 68uH, 47 uH, 33uH, 10uH, 5.6uH and 4.7uH, resulting the best the 3 4.7uH installed on each wire. Wanted to try with 6.8uH but didn´t have on stock, I think this would the best value for its resonant frecuency (37mhz)http://www.ariston.es/esp/detalle.aspx?id=1768

Finally removed tilt twisted wire and installed a regular 3 servo wire with a toroid at the end.

Today I went to my flying field to try all these mods (previously tested on my backyard), the first problem I had was that my OSD have suddenly died, so I reconnected video signal trying to bypass my osd, then I had clear video to test.

I made 2 conventional flights with video on, all went much better than the other day, and I almost have same range than with my previous setup on the easystar, so I think now the installation is acceptable. Since this plane was intended to run with Aerolink 868, I think that the remaining issues are going to dissapear with this frecuency. But if I was going to use 35mhz on this plane forever, the result doesn´t convince me at all, I´m supposed to get more range than with easystar, the components placement is by far better, so results have to be better too. I think that solution goes with different rx, or pan servo, this digital servo is sweet because don´t have to deal with more artifacts to get 180º pan, but for me it has caused lots of headaches.

The video problem have gone away alone, I have done nothing to solve it, I just lowered tiny mic gain, but the other day there were lines even with motor off

Tinymic on its location continues to get over saturated with wind, even with the gain to the minimum. Mr.RC-Cam, you have you mic on the same location than mine, don´t you have this problem too? I would place some foam on it.

All my toroids have 7 tight wraps, btw tubular ones do nothing, and it´s hard to get more than 3 of 4 turns.

So soon I´ll maiden this bird on fpv on 35mhz first and after on 868mhz.

Share this post


Link to post
Share on other sites
Mr.RC-Cam, you have you mic on the same location than mine, don´t you have this problem too? I would place some foam on it.

Here is a test to try: With the mic gain at minimum, speak normally into the mic from a few inches away. The level should be very hard to hear (mostly silent). However, if minimum volume settings still strongly picks up your voice, then I suspect your A/V Tx already has a high-gain mic preamp in it (and Tiny-Mic is not needed in those situations). Otherwise, a piece of foam to act as a wind screen should be used.

All my toroids have 7 tight wraps, btw tubular ones do nothing, and it´s hard to get more than 3 of 4 turns.

I've never been impressed with the tubular variety.

BTW, I'm glad to hear that the glitch issue is essentially resolved for now. :)

Share this post


Link to post
Share on other sites
BTW, I'm glad to hear that the glitch issue is essentially resolved for now. :)

Yes not without your help, can´t wait to maiden in FPV

Share this post


Link to post
Share on other sites

I don't want to get this thread off track but need to ask a ?. Where do you guys get your toroids and what type/size are they? Powdered iron, ferrite, mix, etc.?

Edited by W3FJW-Ron

Share this post


Link to post
Share on other sites
Wanted to try with 6.8uH but didn't have on stock, I think this would the best value for its resonant frequency (37mhz)http://www.ariston.es/esp/detalle.aspx?id=1768

The SRF value is the frequency where the choke has no reactance and essentially becomes a resistor. I don't think you want the SRF to match the R/C Freq. :)

Where do you guys get your Toroids and what type/size are they? Powdered iron, ferrite, mix, etc.?

The best choices can be found at any *real* electronic store (avoid the R/C hobby shop and faux electronic suppliers like Radio Shack). Look for Ferrite type 43 and 77, single piece donut shape. Get several appropriate sizes for experimentation. Basically, something that would allow at least six (more is better) tight cable wraps, yet still be compact. They are very brittle, so after cable wrapping place heatshrink on them for impact protection.

Share this post


Link to post
Share on other sites

Thank you. I'll be visiting Digikey tomorrow.. I should ask are these 72 MC values, or 2.4 gc ?

Edited by W3FJW-Ron

Share this post


Link to post
Share on other sites

The toroids are effective with VHF frequencies.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now