I am working on non destructive testing using hall sensors dc source and i am facing problems in the measurement of voltage.Typically the input voltage is measured input and ground and output voltage between ground and output.The output voltage should be half of input voltage but i am getting it only at one sensors.The voltage of second sensor is not constant.I isolated both the sensors and tested it but it shows the same result.There might be some manufacturing defect in the sensor .I will be using new pair of hall sensors which should be good but i may experience similar problem as in the last setup.There are two hall sensors(a1301-02) (a1308-9) which are used separately in two devices Please refer the screenshots for more information.I wanted ways to reduce voltage fluctuation.There is no connectivity problem nor there are any loose connections.The hall sensor is mounted on a bobbin which encloses steel core and copper coil is wounded on the bobbin.
The schematic diagram for the circuits
Two things to consider before anything else:
First, there needs to be a 0.1uF capacitor place very close to the Vcc and GND pins of each sensor. If it is omitted, or placed on the other end of the cable, it is likely to cause problems. Second, make sure of the pinouts of each part. You did not specify whether you are using the UA package or the LH package. Your diagram would indicate that you have the LH package, but just verify that that is correct.
If all that is correct, there may be issues with noise being picked up in your cable. Try to look for possible noise sources entering the system from there.
Hello David,
Thank you for the reply. The hall sensor package used is ua.i will try testing using the capacitors. I do have capacitors but the capacitance is 100 micro farad.is it fine if I use it or should I strictly use 0.1 micro farad.
I am not really sure about the noise in the cable .The cable I am using is a computer cable which has got some kind of noise reducing cover wrapped around it and some part of it does not have this cable.how do I determine if there is noise in the cable.
Thank you
Have a great day
Is there any restriction of capacitor used like electrolytic capacitor or normal plate capacitor?
Thank you.have a great day
One very important point I forgot to mention previously is that you must be using proper ESD preventive procedures when handling ESD sensitive parts like integrated circuits. This means you are using a properly grounded wrist strap an following proper handling procedures. Otherwise, the Hall sensors can be damaged.
As for the capacitor, there should be a ceramic capacitor placed very close to the hall sensor with the shortest lead lengths possible. As you know from basic electronics, every wire has resistance and inductance, and it also acts like an antenna, which means it can pick up noise from the surrounding environment. Therefore, the shorter those wire leads are, the less resistance and inductance they will have, and the less noise they will pick up. You can keep the electrolytic capacitor there, as it helps keep the supply voltage stable if you have fluctuations in your supply voltage, but the ceramic capacitor is required for good performance
The purpose of the ceramic capacitor is to filter out noise from inside and around the sensor. It essentially creates a short circuit path between Vcc and GND for higher frequencies, and this helps to keep the internal voltage supply (Vcc) stable. Electrolytic capacitors have a much higher internal impedance (both resistance and inductance) due to their physical construction, which make them perform very poorly in filtering higher frequencies. As a note, datasheets for almost all integrated circuits recommend ceramic caps placed close between Vcc and GND for this reason.
Here is an example ceramic capacitor which could be used:
399-9776-ND
Regarding the rest of your physical set-up, the cable looks pretty good, from the image you sent. It appears to be a shielded cable with good connectors on each end.
The more likely place for noise to enter your system would be where you have all of those wires criss-crossing between your cable and the bobbin. Testing for noise is not easy. It involves such things as using an oscilloscope and probing different points on your circuit. When doing so, it is important to remember the characteristics of wire I mentioned above. Because of that, there will be voltage differentials everywhere in the circuit, and the “ground” voltage at one place will not have the same voltage (potential) as “ground” in another place. That is why circuit boards with ground planes are very helpful - the broad area of the plane helps minimize the impedance between various “ground” points on the circuit board, which in-turn, helps minimize noise.
For your set-up, try to minimize the length of those unshielded wires and if possible, twist the wires together which go to each sensor. Twisted wire is less susceptible to picking up noise.
Hey David,
Thank you for the information.i was wondering if the hall sensor which was used for testing is not suitable for measuring the voltage. The hall sensor used was A1309-08 and A1301-2 .They are used for two different setups so one with a1309 and other with a1301.
Thank you
A1301-2-Datasheet.pdf (461 KB)
I’m not certain if I understand your question. If you are referring to what it is designed to measure, it measures magnetic field strength, not voltage. If you are referring to the sensor’s output, it does produce an analog output voltage which is proportional to the magnetic field it senses.
Both the A1301 and the A1309 function very similarly. They have the same pinouts, they both require a roughly 5V supply voltage, and they both output an analog voltage which is ratiometric to the supply voltage and proportional to the measured magnetic field strength. The A1308/A1309 parts are just the next generation of the same type of part with improved electrical specifications. The A1301 and A1302 are now considered “obsolete” and are no longer produced because the A1308/A1309 parts are better parts with no functional differences compared to the older A1301/A1302 parts.
Hey David,
How to determine the ratings of voltage for the capacitor. The voltage supplied to hall sensor is 5 v and the voltage supplied to copper coil is 12 v .so should the voltage should be around 5 v right for the capacitor? It may be a dump question . Sorry for that .I also wanted potential reasons for the setup not working. Like a broken hall sensor output,some problem in the electrical circuit or something like that so that I can fix it.please go through the report where I posted on the forum. There is some part of the cable where shielding is not there but I have covered it with heat shrink tubes.is there any chance of noise entering the cable through that part?
Thank you
Have a great day.
sourabh007,
The capacitor should be rated for higher than the voltage it is in contact with, so in your case, higher than 5V. The 12V should not be directly in contact with it so that voltage is irrelevant. However, there is no harm in going to higher voltage ratings, as that only adds a safety margin so make sure it is not damaged by higher transient voltage spikes.
The capacitor I linked to above ( 399-9776-ND ) is an example of a good option. It is a 0.1uF 50V ceramic capacitor with an X7R dielectric material type. I would recommend using ceramic capacitors which are rated for at least 16V and have either a X7R or X5R dielectric material for your application. See here for further information on this topic:
Ceramic capacitor Temperature characteristics and dielectric classifications
As far as methods of trouble shooting the circuit to try to find the problem, there are a few things to examine.
First, do continuity testing on every pin of the sensor between the sensor and every exposed electrical point all the way back to the opposite end of the cable, and be sure the sensor is getting a solid 5V supply.
If you find a fault in any one of those tests, fix the fault and try the sensor again.
If all of those connections are good, then isolate the sensor from the rest of the system to see if it is operating properly. Once the sensor is isolated from the rest of the system, apply power either directly to the sensor’s Vcc and GND pins, or as close as possible to them (the other end of the wire directly connected to it, ensuring that the wire is not connected to anything else). This way, nothing else can interfere with it. When the sensor is powered with 5V and not sensing any magnetic field (no power applied to the coils), it should output right around 2.5V. If you bring a magnet nearby, it’s output should move up or down from the 2.5V value as it detects the magnetic field. If you do not measure any voltage, or if its output does not vary as you move a magnet nearby, then the sensor chip may be damaged.
If the test directly on the isolated sensor works but your whole system still does not work, then you may have some sort of interference/noise issue. The heat shrink tube you mentioned will NOT help with electrical interference issues because it only protects the wires from physical damage; it has no electrical shielding capability. As I mentioned in a previous response, making those unshielded wires as short as possible and twisting together the wires attached to Vcc, GND, and Vout can help reduce their susceptibility to noise somewhat.
Good luck.
Hey David,
Thank you for the information. I did test the sensor befoe.using all the steps mentioned before and looks like the sensor chip is broken.maybe during handling,maybe during soldering or something.are there any other ways in which the chip might get damaged? I need to justify why the sensor is not working and give reasons for that.
I am also working on li al batteries for thesis. I might need some help later.
Thank you for all the inputs and help.i really appreciate it.
Have a great day
Hi sourabh007,
Damage is most likely to occur during handling or soldering. Handling damage is typically due to ESD. Improper handling procedures or faulty grounding equipment is usually the cause in such cases. Soldering damage is usually a result of excessive heat getting into the package. Using a good no-clean flux and minimizing solder dwell time are the best way to reduce the chance of soldering damage.
If your sensor survived handling and soldering but has subsequently failed, then it could have been damaged by excessive voltage transients or incorrect connections (such as reversing Vcc and GND connections). Transients can occur when power is applied or removed, or when some external noise source gets coupled into the one of the wires connected to the sensor and it causes an over-voltage event. These are more likely to occur the longer the wire is between the power source and the sensor. Placing that ceramic capacitor close to the sensor should help with this. One can also use other means to help protect the sensor, such as using a transient voltage suppressor (TVS) like the SA5.0ALFCT-ND.
If you intend to replace the failed part, I would recommend you test the new part prior to installing, being extra careful to use proper ESD handling techniques (properly grounded table mat, wrist strap, etc.). Make sure you have a quality power supply that will not produce transients when powering on an off. You can place the sensor in a solderless breadboard and put the 0.1uF capacitor next to it. As described above, it should output 1/2 of Vcc when no magnetic field is present, and then that value should vary if you bring a magnet nearby. Once you know you have a good part, then you can install it in your system and go from there.
Hey David,
Thanks for the input. The hall sensors are working fine . But there is one problem . I am getting voltage between input and ground but not between ground and output when the cable is connected to the socket. When I desoldered the output wire and measured the voltage between ground and output ,I am getting voltage. Any idea what is happening? Is it some kind of short circuit? I visually inspected the circuit but couldn’t see any contacts.should I look at something in the data sheet?
Thank you
Have a great day
It sure sounds like you have a short somewhere. Try doing a continuity test between the ground contact and the output contact on the cable connector. There’s a good chance that you will measure a short, or at least fairly low resistance between those two, based on what you have described. If so, then you need to figure out whether the problem is in the cable or some other point between the cable and the sensor. It’s a matter of isolating different sections and testing each one.