J’ai acheté le kit Générateur et capteur de fil de périmètre pour mon projet DIT robot-tondeuse et j’en suis très content ! Cependant avec 120 mètres (390 ft) de fil, le retour du signal est très faible ( quelques Millivolts) sur la carte de réception.
Est-ce possible d’augmenter la sensibilité en modifiant les inductances, remplacement du NE555 , remplacement de l’ampli ou modifier la tension d’alimentation.
Un nouveau Kit avec une longueur de fil périmétrique de 500 ft serait vraiment top !
Merci par avance pour votre réponse.
Bonjour,
Vous pouvez essayer ces deux suggestions:
- Augmenter le voltage d’alimentation de la carte du générateur (jusqu’à un maximum de 15V).
- Pour améliorer les performances du kit, vous pouvez également faire tourner le potentiomètre sur la carte du générateur de manière à ce que vous ayez la distance la plus longue entre l’inducteur et le câble périphérique. Ce potentiomètre fait varier la fréquence d’onde carrée de la carte du générateur de 32 698 Hz à 43 956 Hz, car la fréquence de résonance du circuit de réservoir sur la carte du capteur peut varier en raison des tolérances de fabrication des composants utilisés.
Bonjour.
Merci pour votre réponse. Je vais tester
Si jamais vous développez un projet du même type mais pour “grand périmètre”, n’hésitez pas à m’en faire part. Je suis intéressé !
Cordialement.
Il me semble effectivement que les boucles de robot tondeuse de grande maques soient alimentées en 24 V et permettent une meilleure détection.
Problème réglé ! L’ajustement du potentiomètre - sur site- a réglé mon problème de détection. Avec plus de 120m de câble périmétrique, j’arrive à le détecter à plus de 10 cm . C’est super. Merci
1 Like
I have been having issues with my landroid doing circles intermittently. I brought up the troubleshooting guide and it recommended replacing the perimeter wire. So I’ve completed that. It still does circles intermittently. So here I am. Next step is to order this kit and replace the components. Has anyone had success with this kit on a landroid? Thanks.
It would be better to look for a replacement that is made specifically for the Landroid. As the two systems are most probably different and won’t be compatible to work together.
Hello, I have built the same signal generator but I have an issue. When I check the frequency of the output after setting around 34k, I see that it swings between 33.9k and 34.1k. Is it about the quality of 1uF (C5)?
The frequency of the generator’s output wave can be calculated with this formula :
f = 1.44 / (( R1 + 2 * (R2 + Rpotentiometer)) * C3 )
Therefore, the accuracy of the generator’s output wave frequency will be dependant of the accuracy of the parts above (R1, R2, Potentiometer and C3)
Hi bdaouas,
Thank you for posting this. From your knowledge, if there are (2) different am perimeter wires (a robotic mower and an underground pet fence), is there a way to try to reduce interference if I need to run one across the other at 90 degrees in (2) different places?
I don’t know the frequencies that either of them use and don’t think either allow changes.
If by interference, you mean crosstalk, you shouldn’t have electromagnetic interference issues since your two perimeter wires run one across the other perpendicularly. Crosstalk usually happens when wires are run one next to the other in parallel.
Hi,
I have an issue with this module. I think that @jcf was right. I explain my problem:
On the first module I bought, I had one sensor with good behavior. The second one saturates at maximum value, even when the sensor is far from the perimeter cable.
I have analysed the situation on most of the pins to verify what was happening. I saw that after first stage of amplification, an offset exists and this offset is amplified by second stage up to saturation.
I have bought two new kit. For first kit, the second sensor was working properly, but the first one was saturating.
on the last kit, both sensor are saturating.
I verified all continuity, everything seems to be clear.
The comment of @jcf points a lack of high pass filter between both amplification stage. Unfortunetly it seems to be difficult to adapt the board for that. I intend to test it on a new kit.
I am a bit surprised being the first to have this problem. Could something else explain my problem?
I purchased and assembled an arduino, a motor shield, the signal generator and sensor kit, and a robot chassis with motors and wheels, and I am now developing the sketch and testing the circuits…and I’m new at this. Rather than a perimeter wire, I plan to make this into an inductive guidance/wire following robot. Using the analogread function, I’m getting a scrolling readout same as listed above (400-780), but this raises 2 questions:
-
Why do the readings only fluctuate when the sensor is practically touching the signal wire? I thought this would start fluctuating within 15 cm of the wire.
-
I thought the raw values were to vary from 0 to 1024, according to the guide, but the scrolling example above and my own date show 400-780. What am I doing wrong?
I have not adjusted the potentiometer on the generator since I don’t know what I’m looking at.
THANK YOU!
I answer to myself:
I bought some other quad-op amplifier with the same pinout. I have built one of the module with a support so that i can exchange the amplifier easier.
I have tested .
one other LM324N works very well.
Other amplifier have similar problems. So I realy think that @jfc is totaly right. First stage creats an offset wich is amplified by second stage.
It could be a nice idea to propose an update adding a filter between the two amplification stage.
anyway, thanks for your work !
Hi,
I’m actually facing on the same problem.
I try to figure out how to remove this offset.
Because of this, the output signal of the second stage saturate at 5v.
Any idea of some value of filter between to stage?
Thanks
I am not very experienced in that, but I would say that a high pass filter with a limit at 100hz could solve this issue. This is far enough from the module frequency (>30kHz) and high enough to be sure that any constant value will be canceled.
Before second amplification (5 and 12 on wiring diagram) we have to had a capacitor and a resistor (eelectronique.aop.free.fr/AOP_lineaire_F/Filtre_PHaut_2.html) so that Fc = 1 / (2.pi.R.C)
Yes, I already do this but if I remember it’s not work well. I will test again (I think my previous Fc was about 1kHz) with lower fc.
Yesterday I found another way to do this (in theory) so I will test it too.
I am having the same problem. Were you able to solve the problem? If so, how? Thanks!
Same problem hete. Any news?
Do you have a photo of what the whole thing looks like all together on a robot because I’m confused about where to put everything and what I need so I photo might help