I am using this one: UM982K4 which is similar like holybro UM982.
Understood thanks again. Now I am pretty sure that the glitch came due to power issue. I will power all the electronics separately. I have updated the GNSS system for GPS and GLONASS. Shall I change the Gps rate also? by default it is set as 5Hz.
For the power redundancy, I am going to power the pixhawk by mauch (primary) and power brick (backup). I will power GPS & radio separately. Thanks again for all the clarification.
Hi @amilcarlucas, Thanks for your reply. I am using Mauch sensor and the power cable also looks good. Maybe the mauch is having some issues. Shall I use Pixhawk 6C power brick as the primary power source? My continuous current is around 55A and as per the brick specification it can handle that but I have never used it for high current load application. What is your recommendation?
Understood. Thanks @Yuri_Rage
I am working on it. Btw, what are the parameters should I update to get the best performance from my GPS? I have only updated the GPS_GNSS_MODE to 65, Is there anything else I should update? Just want to add something: With this GPS I have noticed that the Hdop and sat count will remains same which is 0.6 and 28 (on ground and flight). I have never got Hdop below than 0.6 even if i keep the drone for 1 hours in the open clear sky. I have other small GPS they gets the 0.48 Hdop if I keep them for 5 minutes in the open sky.
Hello @scientist_bala I have no expierience with these nice small GNSS modules. But as you ar using the xxxK4 it seems to be the smallesrt one. On the pictures is shown only the GNSS Chip and connectors. So this module will probably not have wide range power regulator and has no capacitors to filter power spikes. Therefore any short power failure will go direct through to the GNSS receiver engine. But without a correct datasheet I only can interpret what is on the small picture seen.
Hi @Juergen-Fahlbusch Sir, Yes you are right It is just a UM982 GNSS chip and connectors.
Exactly, I have been worried about this. Now I am going to power it by 5V 3A BEC, shall I add some capacitor in the setup for the smooth output. What would you recommend for that? (Voltage: 5V and Power: 1.6W)
Energy stored in the capacitor = 0.5xCxV^2 = 0.5xCx(5)^2
Load Resistance= (V^2)/P = 25/1.6 = 15.625
Time constant = Resistance x Capacity = 0.1 (I am considering)
I am not an electronic engineer, So I am really sorry if I miss something or written something wrong. Is that how I should calculate the required capacitor to make the power line smooth for the GNSS? and I guess I need to add some resistor in series to the capacitor so that the current should not damage the GNSS.
Thanks again for all the guidance and help.
Thanks @amilcarlucas for your response I totally agree with you, I have been using it for last 2 years. I followed the wiki instructions for the setup. Btw your multicopter tuning guide is really great and informative. Thanks for sharing it.
Hello @scientist_bala with the separat BEC you are on a good way. From the data you send before it has more than enough power for your GNSS. Without great mathematics you can use a 10yF, a 1yF and a 100nF capacitor all in parallel direct to the GNSS module input. Connected as near to the GNSS module as possible. You don’t need any aditional resistor, just the three capacitorsparallel to the input. The capacitors need to be minimum 10V types and low ESR as possible. Capacitor type ceramic.
Also twisting long lines reduce noise on it.
Hi @Yuri_Rage, I hope you are doing good. I am assembling my drone again based on all of your suggestions. Thank you all so much for that.
To avoid any crash, I am thinking to add M8N GPS in my system as a backup gps. I have read the wiki post, as my both GPS are different (UM982 and M8N) I am not supposed to use blending. I want to use UM982 as a primary and M8N as a backup so for that I need to set GPS_AUTO_SWITCH = 4 right?
and In case if I enable the GPS Blending, will it degrade the performance or it will not improve the performance much? (as both GPS are different). I don’t want to reduce the performance but it is fine if it does not improve the performance and will still act as a GPS redundancy in case of glitch in any GPS.
Thanks again! looking forward to hearing back from you.
Hi @Juergen-Fahlbusch, I hope you are doing great. I have been busy in my academic work lately, now I have again started assembling my drone back.
I am having some issue, my battery connector is damaging while connecting the battery. I know generally some spark comes while connecting the battery due to idle load of the motors but it was not happening earlier (I used the same battery and connector) now sparks comes and connector is becoming little black and damaging. I thought to ask you as I am unable to find the cause.
I am using these capacitors: are they good? and can I use them like this (twist all positive together and negative together and then connecting it in parallel to the GNSS input)? or do I have to use them in some order to each other?
Hi @scientist_bala,
As also @amilcarlucas stated it is better at least for the smalest 100nF capacitor to use not an electrolytic type but a ceramic type. This type reacts on fast spikes better.
To twist the wires as shown on your picture might be good only for a first test but is no solution for your flying machine. On your picture it looks a little bit that the wires can have easly shortcuted. Better you shorten the solid wires and solder flex cables to each capacitor and than to the correct pads somewhere at the GNSS connector. But in principle the three capacitors are all connected together and in parallel to the power input of the GNSS receiver.
When did the sparks come. With or without the extra capacitors?’
That you have so large sparkes that the connector get damaged is not normal. Either it is the wrong connector not good enough for high current but more probably you have a much to great power on current. Normally motors are disconnected during power up by the output stage of the ESC.
Yes, I twisted the wires just to show. I am going to shorten these wires and solder all positive together and negative together and then solder them close to GNSS in parallel.
Without capacitor, I have only added the two 5V bec in the drone and nothing else changed.
Yes! My connector is good enough to handle up to 70A current (I have tested in ground test) and it is only damaging while connecting to the battery while making a spark. I checked the idle current load in the system is only up to 0.5A. Do you also experience this spark generally while connecting the battery?
Hi @scientist_bala,
If you shorten the pins of the extra capacitors and solder them together don’t forget to isolate afterwards all blank metall to protect against any short cut. But I think this is always clear .
Such heavy sparks on the connector that it goes damaged are not usual for me. A spark always is a very high current. So, if ths high current occurse during switch on mostly these are some capacitiv or inductive high loads. Inductive loads are the motors but as mentioned before these should be switched of bey the ESCs. Capacitiv loads are in all kind of powersupplys as BEC, ESC eg. So if you are noticed a change of the sparks you can test which particular component has most influence to these sparks by taking one of the components of during switch on.
If this high current is only during swich on it might be not good but acceptable, but if it occured during flight it we have again a massive power problem as discussed before.
Yes! ESCs and BECs have capacitor, when we connect the battery there is very high current flows through the connector for a fraction of second (demanded by the capacitors) that creates spark and the area where sparks comes is little damaging due to first interaction of high current load.
This is really a good Idea, I will test it. Thanks!
No! after battery connects everything is normal and good. I am just worried about the connector life.
I thing the spark and connector damage (little) is normal and that may be the reason industry drones have switch so after the battery connection you can turn on the switch to power the drone. In this case there will not be spark and no damage.
Btw, have you used soft-start circuit? It minimizes large startup currents from flowing when the input power is first applied.
Thanks @amilcarlucas but I am using blade connectors as I have assembled my battery pack on my own, to use this I need to again change my design. Just looking for other alternatives to solve this problem otherwise at the end I will try these.
These days capacitors are higher quality than they used to be - there is not so much need for the low value/high frequency caps like 100nF or less. The ones already in place will do that job anyway.
Generally you just need an electrolytic or tantalum cap of the highest value you can physically fit in the available space.
Tests I’ve seen of decoupling and filter capacitors even got good performance out of a 1000uF instead of all that bunch of smaller to larger capacitors in parallel.
So one good cap of 47uF or up to 100uF would work well with small devices depending on how much variation and noise you are seeing on the supply voltage. Keep it as close to the load as possible.
A 47uF tantalum is physically quite small.
Hi @scientist_bala
What kind of blade connector you are using, Can you give details / image?
On my Helis I use only max 3S-800mAh batteries, so there is no need of soft start.
Hi @xfacta, thanks for your response and it is good to hear from you again.
Yes! I have checked it, it is quite small and can be fit easily. I will try that.
I am using like these connectors
My battery is quite big and powerful. I am thinking to use a switch so once battery is connected, I will turn on the switch to complete the circuit. Just looking for high amp switch which will be good for my use case.
Hi @scientist_bala, you are joking.
The connectors following your link are only specified by 7A per pin. Can you show a picture from your real setup.
Add a additional switch might add other problems. Also if all soldering ist done very well and you are using a very high quality switch you add some resitance and power loss inline.
Keep the system mainly the power line s simple and robust as possible. So I recommend to change to a proofed connector system for this application like AMILCARLUCAS wrote before.
.