Pololu Robotics & Electronics
Menu
My account Comments or questions? About Pololu Contact Ordering information Distributors

Adjustable 4-12V Step-Up Voltage Regulator U3V50ALV


#1

I am trying to use a U3V50ALV to step up the voltage from a 4 cell NiMh battery pack to 6 volts for an RC receiver and two servos. I know from load testing without the regulator that when I stall both servos I will pull roughly 1.75A. Based on the efficiency curves it seems like this is in the range of what can be expected from this device.

When I put the regulator in the system however it does not seem to be able to maintain a 6 volt output even under relatively light loads so I am wondering if I am doing something wrong or if I have chosen the wrong product? The wiring is soldered directly to the board.

I’ve listed some data below in case it’s helpful (the batteries were partially discharged to start with):

Current A (out of reg) / Vbatt / Voltage out of reg / comment
0.05 / 4.86 / 6.06 / no load on servos
0.27 / 4.65 / 5.86
0.38 / 4.52 / 5.71
0.52 / 4.39 / 5.51
0.80 / 4.13 / 5.08
0.94 / 3.97 / 4.87


#2

Hi.

I am sorry you are having trouble with your regulator. Could you give more information on your battery pack? If you fully charge it, does the problem get better? Have you tried testing the regulator with no load or with just the receiver? Can you post pictures that show your whole setup and all connections?

-Claire


#3

Claire,

Thank you for your reply. The same issue exists if I use fully charged batteries. The battery pack is made up of four 2600mAh AA NiMh cells in series. Once I start applying a load the output voltage from the regulator starts dropping. At full load (i.e .servo stalled) with a fully charged set of batteries the battery voltage drops to about 4 volts and the output from the regulator drops to just under 5 volts. These are approximately the same numbers that I got with the partially discharged set of batteries previously (although because they were not fully charged I reached those voltages at less than full load.)

The first image shows the regulator. I have a standard RC servo type of connector on each end
The second image show the back of the regulator with the solder connections
The third image shows the installation but, since it’s a bit messy to just look at in a photo…
…the fourth image shows the circuit diagram

Am I mistaken in assuming that the regulator should maintain (more or less) a constant voltage on the output in this application?

P.S. I don’t seem to be able to successfully post/attach photos to this message - perhaps because I am new to the forum? I have posted the four pictures referenced above to a dropbox at this link.

P.P.S. I forgot to reply explicitly to one of your questions. The regulator outputs 6V under no load conditions and I can adjust it with the pot as expected.

John


#4

Hi, John.

New users should be able to post or attach up to 50 images and the max file size is in the megabytes, so you should not really be restricted from posting pictures. Please let me know if you have any further trouble with that.

The regulator should maintain the set output voltage as long as the load on it is not too high. From the pictures, your setup looks okay and those batteries seem like they are appropriate, so I suspect either the load of your servos is larger than you think or the regulator is having trouble dissipating heat. How did you measure the current draw of your servos? Could you link to a datasheet or product page for them? Where are you measuring the output voltage? If it is not already directly on the soldered pins of the regulator, does measuring there change the readings?

-Claire


#5

Claire,

Unfortunately I do not have a link for the servos because they are generic servos that came with the rc sailboat that this all relates to. I originally measured the current draw with a DVM/ammeter and also with a watts-up power meter and both gave the same result (i.e. around 1.75-2A) so I am pretty confident in that number. The output voltage is measured at the receiver but when I check it at the regulator it shows the same values.

I obviously don’t know the product as well as you do but I don’t think it’s a heat related issue because 1) the voltage droop is immediate with load and 2) I’ve left things under load for several minutes (simulated sailing so to speak) the regulator is warm but not too hot to touch anywhere on it.

Is there an alternative test I can run on the regulator completely separate from the receiver and servos, maybe just connect it to the battery pack and put a 5ish ohm resistor across the output?

John


#6

Testing the regulator with a purely resistive load would be good since it will be easier to know for sure the current draw. Also, do you have another power supply you can test with? The capacity spec of your battery pack seems good, but the voltage drop seems pretty high.

-Claire


#7

Claire,

With a purely resistive load I got the following data with freshly charged batteries:
Load = ohms load at output of regulator
Vin = Voltage at input to regulator
Iin = current at input to regulator
Vout = Voltage at output of regulator

Load/Vin/Iin/Vout
open circuit/5.79/760micro amps/6.06
13.1/5.24/0.59/6.06
6.7/4.45/1.40/6.06
4.4/3.03/3.0/5.56

Thus with a purely resistive load the regulator appears to behave as expected. It maintains voltage until the input voltage drops to around 3 V (I think the spec is 2.9V?) and when you run the numbers the efficiency based on power out divided by power in is around 85-90%. I did notice that at the 3.0A Iin current level there is a large voltage drop from my battery to the input of the regulator (around 1.5V, not reflected in the data above, I didn’t have enough DVMs to make all of the measurements simultaneously) so I need to check on why that is - maybe just the relatively small gauge of the wires).

I then took the same (now partially depleted) batteries and connected to the receiver and servo as I have in the past. I got the following data point when I stalled the servo;

Vin = voltage at input to regulator
Vout = voltage at output of regulator
Iout = current output from regulator to receiver and servos

Vin/Vout/Iout
3.65/4.40/1.42

In the end I’m still confused. I can understand the regulator output falling when the input voltage drops too low (as it did at with the 4.4 ohm load in the resistive test) but when I connect to the receiver and servos the input voltage appears to be staying high enough (3.65V) that it doesn’t seem like that should be an issue. Also, as the original data shows, when used with the receiver/servos the regulator output voltage won’t maintain the setpoint even at very low output current levels.

I am going to see if I can scrounge up a power supply to get the batteries out of the equation and if I can find a suitable one will collect more data. Any thoughts you have in the meantime are welcome.

John


#8

From your latest testing it seems clear that the load of the servos is key to the issue. When you measured 1.42A from the servos and receiver what were the servos doing? Can you try removing the servos from your sailboat and just connecting one of them? Multimeters often average several samples over a relatively long time and servos can draw current in bursts when they are changing position and straining, so it is possible that you are missing larger current spikes. Can you look at the voltage on the output (or better the current of the output) with an oscilloscope?

-Claire


#9

You might also try adding some large capacitors to the regulator output to see if that helps.

-Claire


#10

I agree that the servo load is somehow being an issue. The 1.42A was measured with one servo stalled and the other doing nothing. In sailboat terms the sail servo was pulling hard against the wind trying to hold the boom in place (and was just breaking even so to speak) while the rudder servo was neutral and unloaded.

I do have a small handheld oscilloscope that I can use to look at the output voltage but I don’t think I have a current probe.

I can also try some capacitors although in the real world (ok, it’s an RC “real world” :grinning:) I imagine they would discharge relatively quickly if you are exceeding the ability of the regulator to hold voltage? Any suggestions on a starting value?


#11

A few hundred micro farads is probably a good place to start.

-Claire


#12

Claire,

The attached image shows data I captured with my oscilloscope. I put a 1 foot length of solid 26 gauge wire (0.016 in. diameter) in series with the positive lead to the sail servo . According to the internet this size wire has a resistance of 40.8 ohms per 1000 feet so a 1 foot length should have a resistance of 40.8 milliohms. I used a constant current power source to measure the resistance of this particular piece of wire and got 48 milliohms which I would consider pretty good agreement all things considered. The scope trace is the voltage drop across this 1 foot length of wire (0 volts is at the first division up from the bottom, where the small arrow is on the left side).

In any event, you can see that the average peak voltage drop across the length of wire is around 80mv which, using the resistance values above, translates to around 1.6-2A (depending on which resistance value you use) which is consistent with what I have measured with the DVM previously. The occasional short-lived spike of around 100mv translates to 2-2.5A but they are maybe 1-2 mS wide and are not very frequent so it’s hard to imagine they are causing the issue. While this data was captured the input voltage at the regulator on my DVM was at 3.4V and the output voltage at the regulator was 4.9V. These batteries were partially discharged so I also tried with fully charged batteries and nothing really changed - the input voltage at the regulator went up to 3.5-3.6V but everything else stayed the same.

I also tried capacitance across the regulator output (100 micro farad and then 470 microfarad electrolytic) and it doesn’t make any difference.

What do you think? I’m not sure where else to go/what to try and I still don’t understand why the regulator is not holding 6 volts at the output as it does not seem like the current being drawn by the servo is excessive based on the regulator specs. Is there reason to think that there is an issue with this particular regulator in your experience? Is there another that is of similar size (I don’t have a lot of physical room) that has a higher capacity or is there an alternative solution that you can think of? I’m simply looking for a way to use the rechargeable cells in the boat instead of disposable alkaline cells while still maintaining the higher voltage to get better torque on the servo. EDIT - SEE NEXT POST ALSO

Thanks.

John

P.S. I looked up the specs for a servo similar to this one (one that people use to replace this one if it goes bad) and according to the manufacturer the stall current is 2.5A so again consistent with the numbers I’ve been seeing.


#13

Claire,

One more set of thoughts after sleeping on things.

In looking at the above data and the efficiency curves on the product page it is clear that my system is operating pretty much on the curve for 3.3V input/6V output. Given the output current that I’m drawing at stall (1.6ish amps) it is pretty clear that I’m at or just past the end of the that curve. Thus my conclusion is that the regulator is probably behaving like is should but the output is just “falling off the cliff” like the curve shows and I’m simply limited by the efficiency/my input voltage.

I don’t yet understand why my input voltage is getting dragged down so far because when I measure the current directly out of the batteries I am getting around 2.5A (i.e. not quite 1C) so I’d expect them to hold voltage better than that. As a comparison I also tried a bench top power supply set at 4.8V and it got pulled down to 4V @2.6A which I don’t understand either.

I need to explore the above further but assuming that I am stuck with the current conditions I don’t see any other boost regulators that you sell that seem like they would work. Products with lower allowable input voltages (like the 791) don’t appear to have the input or output current capacity I would need. Would you agree with that conclusion and/or can you think of any other solutions that Pololu has that I am missing?

Thanks.

John

P.S. Another idea/question: Is there a way to make two or more boost converters “play nice” if wired in parallel to increase the output current capacity? I would think that you have to have some sort of load equalizing circuit or small differences in performance would lead to one converter taking most of the load and leading to burn out…?


#14

I agree that, given the current you measured with the scope and the 2.5A stall of the similar servos you found, the load of your servos is just too much for the regulator given your input voltage. If you are seeing 1.6-2A draws on just one of your servos, it seems unlikely to me that you would be able to power two servos off the regulator even with a steady 5V source. The efficiency curves of the regulators show data out to the maximum current possible for a few seconds, and the maximum continuous current output is typically at least a few hundred milliamps lower. For the U3V50ALV regulator with 5V in and 6V out, I would expect to get around a little over 2.5A continuously.

Unfortunately, we do not have any higher power regulators that would work for you and we do not recommend paralleling regulators. If you are interested, this thread has more details on the issues with it. If you are able to figure out the issue with your input voltage dropping, you might consider powering each servo off of separate U3V50ALV regulators.

-Claire


#15

Ok, thanks for your time Claire. I will certainly explore why I am getting a larger voltage drop than I expect and I’ve seen some information elsewhere about using transistors to make a current mirror to balance the load between two boost amplifiers in parallel which sounds potentially interesting too (recognizing that it isn’t necessarily an official Pololu sanctioned solution :grinning:). Regardless of the outcome I’ve still learned/am learning a few things along the way so that’s worth something in and of itself!

John


#16

If you do try out a current mirror or some other circuit to manage paralleling some regulators, we would be interested to hear about it!

-Claire