Voltage Regulation

Thread Starter

Stillers48

Joined Jul 12, 2020
21
First off, let me clarify that my experience is somewhere in between Beginner and Primate. This is my first adventure into building my own IC and Im sure that my post will be full of wrong assumptions and misused terms. Go gentle on me. That being said, Im currently in the early stage of designing a customized Power Distibution and Control board for LEDs because no one seems to make what I want. It is full of hodge podge parts that are meant for other uses, or wont do what I think theyll do at all. Any suggestions or pointers outside the scope of my question are definitely welcome. Especially in regards to the power system and voltage regulation. At any rate, I have a question about Power regulation within the circuit. Ive included my two preliminary renderings Im working on, the primary circuit diagram however only shows the power system so far, not the components behind it. havent gotten that far yet. The power will come from two NiMH batteries in parallel rated, theoretical, at 7.2V/4400mAh.. All of which is going to be delivered to the circuit components through 3 UBEC Step down chips to get the voltage to the appropriate level for the components and lights. The UBECs regulate power to 5V/3A. Each UBEC is intended to account for one of the three seperate "systems" I plan to build. My question is this:

1. At least one component will draw constant power of 5V/ 2,5A with occasional, momentary surges of up to 4A which is well under the UBECs supplied current. However, with the UBECs set up "In parallel" so to speak, and all rejoining into a single circuit before being diverted back out to each individual system Im going to build, when the current surges above the individual UBECs current limit, will the component be able to meet its current demand by pulling the additional current through another UBEC? For example, Component A surges to 4A, to meet the demand, 3A channels through UBEC 1 and an additionjal 1A is drawn through UBEC 2? and all components.

2. Considering the pathways that the UBECS are channeled and then all supplied power being repooled before being applied, what will be the theoretical levels of power downstream of the UBECs? Does power/.amperage follow the same rules as batteries when being wired in series or parallel? Or is it stable at the levels set by the UBEC?

3.. Do I have to return the ground for each circuit to the UBEC after application in the circuit? Or am I able to ground the end of each circuit to one of the downstream ground circuits? And if I can, when the back end of the circuit is grounded to the circuit ground pathway, will it naturally reroute back to the UBEC even if the pathway is shunted directly back to the battery negative? Or will it bypass the UBEC ground path and "head on home" to the battery negative itself, if the pathway is open to do so?

Im also him-hawing with switching the power supply over to 4 3.7V 3500mAh cell phone batteries in series to get the circuit up to 12V to make possible using a few different components. As id still be using 5V components and using the UBECs, Im curious, what happens to all the scrubbed off voltage in that scenario? Any opinions and suggestions in general about the batteries, power distribution, configuration, UBECs, or breadboard power supply circuts would be greatly appreciated. Seriously, take me to school. Thanks!

Tim
 

Attachments

Papabravo

Joined Feb 24, 2006
14,245
Normally, connecting power supplies together is a problematical enterprise. Since you don't know the state of charge for your primary batteries, you can't predict the behavior when when one is discharged more than the other. What can happen is called current hogging making the problem worse. The typical solution is to use diodes to connect the batteries to the UBECs. Each UBEC can now pull current from either battery. If all the UBECs demand more current than the batteries can provide, the battery voltage will drop, and the UDEC output voltage will drop as well. If your regulators are designed to tight margins they may drop out of regulation increasing the current demand of the UBEC loads. A partial solution to this problem is the use of a buck-boost switching regulator. You should also anticipate an over current sensor to shut things down when they start to go South.

If you have not already done so, I highly recommend you visit Battery University for an education in the care and feeding of batteries.
https://batteryuniversity.com/
https://batteryuniversity.com/learn
 

dl324

Joined Mar 30, 2015
11,288
This is my first adventure into building my own IC and Im sure that my post will be full of wrong assumptions and misused terms.
IC usually stands for Integrated Circuit. You won't be designing one any time soon.
The power will come from two NiMH batteries in parallel rated, theoretical, at 7.2V/4400mAh.
A 7.2 volt NiMH battery is 6 cells in series. Series is already bad enough, but to parallel two packs just increases potential points of failure. You'd be better off using some Li-ion batteries so you'd only need 2 in series.
3 UBEC Step down chips to get the voltage to the appropriate level for the components and lights.
I had to google UBEC:
clipimage.jpg

I guess you mean Ultimate Battery Eliminator Circuit; although I don't know about the use of ultimate.
 

Papabravo

Joined Feb 24, 2006
14,245
I had to Google UBEC as well, but I got Universal Battery Eliminator. We are obviously not tapped into the "Drone" community. I knew NiMH batteries were problematical, but I could not remember why. That is why I suggested a brief sojourn at Battery University. It will quickly rip off the rough scab of noobness. In drones, as in full size aerobatic platforms, you want the biggest possible engine on the smallest, strongest, and lightest air frame. Flying a Pitts Special was one of my thrills of a lifetime.

https://en.wikipedia.org/wiki/Pitts_Special
 

Thread Starter

Stillers48

Joined Jul 12, 2020
21
Normally, connecting power supplies together is a problematical enterprise. Since you don't know the state of charge for your primary batteries, you can't predict the behavior when when one is discharged more than the other.
I figured there would be more to it than Id thought, To be honest, I spent all my time thinking about the downstream effects of multiple power supplies. I never once thought about what was happening upstream.

The typical solution is to use diodes to connect the batteries to the UBECs
This has been a primary concern of mine from the beginning about the overall build. This project has extremely limited space and weight limitations. From the beginning Ive been worried about having to add in this or add in that to get it to function correctly. Im limited to a space of 60mmX90mmX25mm in which all wires, components, and batteries must fit. at present the batteries alone consume about 70% off the space and 90% of the weight cap. This is why I am thinking about the switch to the Lipo batteries. I could get 25% more current, 80% more voltage and decrease the space im using by nearly half. Even more if I stayed at 7.4V and didnt increas to 14.8. The flat design eliminates virtually all the dead space that the Ni-MH's create.

If all the UBECs demand more current than the batteries can provide, the battery voltage will drop,
If you have not already done so, I highly recommend you visit Battery University for an education in the care and feeding of batteries.

If i did my research correctly, the components should, at no time, demand more voltage or current than the battery has capacity to provide. Itd get very close. but not providing for unforseen spikes or factors leading to an increase of current, the theoretical max possible current surge should never exceed 4000mAh. The battery should never be exceeded. only the limit of each individual UBEC.. IN THEORY.......

A partial solution to this problem is the use of a buck-boost switching regulator.
I had seen these and light research did suggest to me that it was a better alternative for the UBECs. I ultimately decided against it, admittedly without doing enough research into them, due to the space they take up. the models I seen would have taken up nearly all the available surface space I have. Id have stack thje component ciruit board on top of it and that is an impossibility for the Ni-MH driven design. In the Li-Po driven design it could be possible, but that may only be true if I remained at 7.4V. Bearing in mind unforseen events and unknown future needs, Id really like to at least keep the possibility of 12V power open to me.

You should also anticipate an over current sensor to shut things down when they start to go South.
A legitimate concern. One of mine from the beginning as well. My decision to not include any failsafes was not without reason. Admittedly, the logic may be flawed or misguided, but from my wheelhouse, made sense at the time. All the components downstream oif the batteries are either Led lights or cheap, easily replaceable circuit chips. I saw very little hazard of fire or explosion from anything other than the batteries. And don't expect to use the system much more than 15-20 minutes at a time. It was to my understanding that lights didnt suffer from voltage or current over/underages in the same manner as other electronic components. They would either be brighter or dimmer with perhaps a shortening of TOL depending on how badly their limits are exceeded. Dependent on frequency of loss, the cost of component replacement due to failure from abuse should be negligible so opted instead to not install falilsafes to conserve space. In order to protect against the risk of fire or explosion from the batteries I decided to design the enclosure so that the batteries are underneath the circuit with a 28 gauge plate integrated into the enclosure between the batteries and the circuit, as well as the drone itself all above the plate. and create an airtight enclosure for the batteries. I also plan to set up the 3D printer to make the bottom side of the enclosure that the batteries rest on to be ultra thin. Thin enough that any sort of excessive heat created by them will result in the bottom becoming structurally unstable to support the battery weight and literally let the battery "fall through" the bottom and off the drone. And because the space is air tight, that design allows it to become a burst panel should there be any sort of explosion inside the compartment. The concussion of an explosion will result in the panel being blown off and falling away from the drone. IN THEORY....

To be dead honest, Im beginning to realize that every time i "bump my head" on this project its always, ALWAYS related to 12 Volt power. Or to needing multiple batteries to meet power requirements. Clearly I should just eat that bullet and build a stable system that uses alll its parts as intended. Im just hoping that before I abandon my expanded plans that maybe I could find some work arounds or better ideas from people that are in the know.

If you have not already done so, I highly recommend you visit Battery University for an education in the care and feeding of batteries.
I will most certainly do this! Thank you very much for all of your input!!!
 

Thread Starter

Stillers48

Joined Jul 12, 2020
21
IC usually stands for Integrated Circuit. You won't be designing one any time soon.
Ahhhh. I. C. I also stand corrected. Thank you.

A 7.2 volt NiMH battery is 6 cells in series. Series is already bad enough, but to parallel two packs just increases potential points of failure. You'd be better off using some Li-ion batteries so you'd only need 2 in series.
It was to my understanding that Li-Ion batteries functioned better the thicker or bigger they were whereas Li-Po batteries thrived as they got thinner. Is this an incorrect assessment?

I guess you mean Ultimate Battery Eliminator Circuit; although I don't know about the use of ultimate.
I was under the belief that it stood for Universal. But, in this world, I normally find myself standing in quick sand. Their normal usage in an electronic circuit would tend to suggest that Ultimate does fit.
 

Papabravo

Joined Feb 24, 2006
14,245
I figured there would be more to it than Id thought, To be honest, I spent all my time thinking about the downstream effects of multiple power supplies. I never once thought about what was happening upstream.



This has been a primary concern of mine from the beginning about the overall build. This project has extremely limited space and weight limitations. From the beginning Ive been worried about having to add in this or add in that to get it to function correctly. Im limited to a space of 60mmX90mmX25mm in which all wires, components, and batteries must fit. at present the batteries alone consume about 70% off the space and 90% of the weight cap. This is why I am thinking about the switch to the Lipo batteries. I could get 25% more current, 80% more voltage and decrease the space im using by nearly half. Even more if I stayed at 7.4V and didnt increas to 14.8. The flat design eliminates virtually all the dead space that the Ni-MH's create.



If you have not already done so, I highly recommend you visit Battery University for an education in the care and feeding of batteries.

If i did my research correctly, the components should, at no time, demand more voltage or current than the battery has capacity to provide. Itd get very close. but not providing for unforseen spikes or factors leading to an increase of current, the theoretical max possible current surge should never exceed 4000mAh. The battery should never be exceeded. only the limit of each individual UBEC.. IN THEORY.......



I had seen these and light research did suggest to me that it was a better alternative for the UBECs. I ultimately decided against it, admittedly without doing enough research into them, due to the space they take up. the models I seen would have taken up nearly all the available surface space I have. Id have stack thje component ciruit board on top of it and that is an impossibility for the Ni-MH driven design. In the Li-Po driven design it could be possible, but that may only be true if I remained at 7.4V. Bearing in mind unforseen events and unknown future needs, Id really like to at least keep the possibility of 12V power open to me.



A legitimate concern. One of mine from the beginning as well. My decision to not include any failsafes was not without reason. Admittedly, the logic may be flawed or misguided, but from my wheelhouse, made sense at the time. All the components downstream oif the batteries are either Led lights or cheap, easily replaceable circuit chips. I saw very little hazard of fire or explosion from anything other than the batteries. And don't expect to use the system much more than 15-20 minutes at a time. It was to my understanding that lights didnt suffer from voltage or current over/underages in the same manner as other electronic components. They would either be brighter or dimmer with perhaps a shortening of TOL depending on how badly their limits are exceeded. Dependent on frequency of loss, the cost of component replacement due to failure from abuse should be negligible so opted instead to not install falilsafes to conserve space. In order to protect against the risk of fire or explosion from the batteries I decided to design the enclosure so that the batteries are underneath the circuit with a 28 gauge plate integrated into the enclosure between the batteries and the circuit, as well as the drone itself all above the plate. and create an airtight enclosure for the batteries. I also plan to set up the 3D printer to make the bottom side of the enclosure that the batteries rest on to be ultra thin. Thin enough that any sort of excessive heat created by them will result in the bottom becoming structurally unstable to support the battery weight and literally let the battery "fall through" the bottom and off the drone. And because the space is air tight, that design allows it to become a burst panel should there be any sort of explosion inside the compartment. The concussion of an explosion will result in the panel being blown off and falling away from the drone. IN THEORY....

To be dead honest, Im beginning to realize that every time i "bump my head" on this project its always, ALWAYS related to 12 Volt power. Or to needing multiple batteries to meet power requirements. Clearly I should just eat that bullet and build a stable system that uses alll its parts as intended. Im just hoping that before I abandon my expanded plans that maybe I could find some work arounds or better ideas from people that are in the know.



I will most certainly do this! Thank you very much for all of your input!!!
I see and feel the pain of trying to get everything right on the first go around. I spent half a century designing and building products for a living, and I discovered a couple of immutable lessons:
  1. Be prepared to throw the first one away, because you're going to end up doing that regardless of any other forces, influences, or considerations.
  2. First make it work, then make fast, small, cheap or adjective of choice. Last make it elegant.
  3. We learn more from our failures, than from our successes.
  4. Fail quickly. Never engage in slow stupid failure.
 

Thread Starter

Stillers48

Joined Jul 12, 2020
21
I see and feel the pain of trying to get everything right on the first go around. I spent half a century designing and building products for a living, and I discovered a couple of immutable lessons:
  1. Be prepared to throw the first one away, because you're going to end up doing that regardless of any other forces, influences, or considerations.
  2. First make it work, then make fast, small, cheap or adjective of choice. Last make it elegant.
  3. We learn more from our failures, than from our successes.
  4. Fail quickly. Never engage in slow stupid failure.
I believe I solved alot of problems today. I purchased 3 5000mah 3.7V Li-Po batteries that are only 3mm thick each. Im not gaining much in terms of current, just a 600mAh gain, but the significant gains are in the increase to 11.1v the the MASSIVE decrease in space dedicated to batteries alone. Now the voltage is close enough and I have the space that I believe I can stack a power supply in with the circuit board and regulate it uo to 12V/5A. The biggest possibility this opens up is potentially doubling my flight time. The drones primary battery is 11.1V 4500mAh Li-Po battery . I get 32 mins of flight time from it. Unfortunately it will take a month or so for the batteries to arrive but thats ok. Im gonna need that month to head back to the drawing board and figure out how I can shunt those batteries into the drones primary power circuit when im not using them to power the light array. Now that, will most certainly not be done with any corners cut.
 
Top