TM4C123GXL: How to power from a battery

Hi,

The input at TPS73633 regulator is +5V and then the output is +3.3V.

There is a VBUS Socket at bottom left of Tiva Launchpad, where you can connect a external power source. See, post below. Other E2E members can give expert advice regarding connecting 3.7V Lipo to VBUS.

- kel

Zero_PD said:

Can I power from a 3.7V Lipo?

That's a good (high, multi-user appeal) question & I'll attempt to guide via a "method" as opposed to "cookbook."

• You do "not" want to introduce any external voltage source to the output of your board's 3V3 Regulator.   (unless you've defeated the connection path from that regulator to the board's 3V3 bus.)
• Find the spec sheet for your board's 3V3 Regulator - revealed therein should be the acceptable, "Input Voltage Range" for that Regulator.   Such will determine if your 3V7 battery meets the regulator's spec.   (my suspicion - especially under heavy load - is that it will not (3V7 is too low))
• Further - won't your battery start decreasing in voltage under operation - thus may "pass" from "in spec" to "out of spec!"

You (and others) may note that most 3V3 linear regulators have a rather limited, input voltage range.   Such confounds battery usage - such as yours.  

My small tech firm usually - but not always - will seek a, "Wide input voltage range" 5V0 regulator - capable of 750mA (or higher) output current.   And - far & away - most such "modern" regulators are of the switching type.   While more complex - these ARE the "future" - and the switching efficiency of these newer regulators eases the "power dissipation" which a linear 5V0 regulator must sustain - should your input voltage be more than a few volts beyond the 5V0 output.  

Normally we "stylize" such regulators - building upon a compact "adder" board - which is easily mounted (or cabled) to the target (5V0 seeking) board.

Once you've achieved the "holy grail" of:

• wide input voltage range
• adequate current output
• 5V0 regulation

such can be directly injected into the 5V input of your board (often via a USB connector) - well satisfying your board's 3V3 Regulator - yet (now) enabling a far wider range of input voltage...    And w/out "any" (warranty voiding) modification to your Eval or Development board!

Note further - certain of these switchers will enable input voltages of LESS than the 5V0 output - thus offering the ultimate in input voltage flexibility...

Hi Amit,

For o.p.'s benefit - can we agree that poster's "feed" of a freshly charged 3V7 battery is unlikely to maintain full/proper 3V3 output - especially as the battery drains - and output voltage flirts w/3V3 and then drops below?

This "operation from a battery" AND the "far wider input voltage range" led to my suggestion of a "switching regulator..."

Again - your eval/dev/LP board's 3V3 regulator remains intact - my suggestion adds an (external) switching regulator - which provides 5V0 output voltage - to be input to your eval/dev/LP board.   Such adds immensely to the input voltage range - especially so when battery powered...

Thank you guys for your input, this has been interesting. So, ignoring the VDD pins on the board and concerning ourselves with the VBUS pin, this seems to be the consensus:

VBUS is connected to an internal LDO (I had to look this up, but it stands for low dropout) voltage regulator, the TPS73633DRB. The DRB takes an input from 3.5V to 5.5V and outputs 3.3V. Can you power VBUS from a 3.7 LIPO? Yes. But it is a questionable decision because you will soon drop below the 3.5V which is the minimum full-rated performance input voltage to the DRB.

Two alternatives remain. Find a 5-5.5V battery and hook that directly to VBUS. This is tricky because no standard LIPO, NiCd, NiMh, or even SLA batteries come in this range. This leaves us with our final alternative and probably best option: regulate a higher voltage battery down to 5V using an external regulator, and then feed that voltage into VBUS. Two-cell LIPOs (7.4V), NiCd and NiMh (7.2V), and 6V SLA batteries all fall into this range. I'll personally use a 7.4V LIPO and a basic L7805 Voltage Regulator, because my circuit is drawing little current.

Does this all sound correct?

Hi Amit - I don't quite understand. I'm afraid we're running up against my knowledge of embedded systems!

1. What do you mean when you say VBUS is driven from the Host? What is the Host?
2. According to the schematic in the tiva launchpad user's guide, the VBUS pin is routed to an LDO voltage regulator that is built into the launchpad board, right? So how would I break this connection?
3. Will putting 5V on the VBUS pin correctly power the tiva board?

Hi Amit - I haven't been able to find a labeled Tiva picture in any of my documents that references these pin names. The closest of I've gotten is this image from the User's Guide. Is "SW3 Switch" the Power Select Switch? And which of these components is the H18 Header?

And - if you insist upon "old sKool" 7805 - be mindful that such regulators (usually) drop at least 2 volts when under load. (sometimes more)

Thus - sad to say - your "improvement" over the (suggested) switcher's WIDE input range - has bitten you - again!   (i.e. your 7V4 will soon drop sufficiently that you'll not long "hold" your 5V0!)
 
Switchers rule - tired, old, lossy "HDO" regulators - not so much!

That VBus pin question is vendor specific - best left for Amit. (my firm uses ARM MCUs from multiple vendors)

Review the spec for 7805 - you'll note the rather large "drop" which rises w/current. We've not used such in years - but our (past) rule of thumb was to always (and only) feed 7V5 (minimum) w/7805 - to insure that we regulated to 5V0!

Note further that should you provide 7V5 input V - and draw the 200mA you state - you'll dissipate 0.5 Watts across that VReg.   Can it sustain that?

This is certainly more complicated than I thought it would be. I am trying to send out a PCB by the end of the day and I just wanted to double check that my external powering was correct. It seems that instead of a simple answer I've fallen down the proverbial rabbit hole.

I have looked through the Battery Booster Pack design trying to understand at which point it interfaces power with the tiva launchpad. It is still hard to figure out an answer. For instance, Section 3.3.1.1 (Launchpad Jumper Settings) states "Ensure that jumper P1.0 is closed and P1.6 is open on the MSP430 LaunchPad." followed by a picture of the launchpad. This picture is unlabeled and I have no idea which jumpers p1 and p1.6 it is referring to.

So I skipped to the Schematic to see if I couldn't figure it out. I payed close attention to the two Regulators at the top of the schematic. The 5V regulator on the right takes in voltage from the LIPO battery pack and outputs to 5V to JP1 Pin 2, which is definitely the VBUS pin. Seems like this is the answer. However, the 3.3V regulator on the upper left side of the schematic takes in voltage from the LIPO Battery Pack as well, and outputs 3.3V to JP1 Pin 1, which I know is a 3.3V output pin. So I find myself still unsure.

I'm about to take this all to my professor. He's designed carrier boards for the tiva launchpad before, maybe he'll have answer for how on earth I do this...

Zero_PD said:

I'm about to take this all to my professor.

But - with the future (already) here - why must we wait?

Most engineering - both now & in the (real) future - involves trade-offs and balancing.   (and same holds true in law & contract negotiations)

Absence of complication may not be a wise (or proper) objective - instead may indicate that the issue is "too simple" or has not been properly investigated - and/or analyzed.

You are not completely innocent in this "bout w/complexity" - are you?   Is not your desire to add battery power - which strays from the board designer's (central) usage/intent - a major component of such "complexity?"

As long suggested - a proper, switching regulator - able to output 5V0 @ adequate current - may be used repeatedly.  (due to the fact that 5V0 has enjoyed long & vast usage - and is accepted by most all (newer) 3V3 regulators.)    And this serves as a powerful "complexity reducer" - even when - and especially when - the future is (really) NOT yet here!

A proper switching regulator sounds fine. My only question at this point is where you input that proper 5V to. Does that 5V go to VBUS? A header on the board? Therein lies the question.

Edit: Just checked with my professor, who designed a protection/breakout board for the tiva launchpad. 5V on the VBus line is what you need to power the tiva launchpad. So yes, I'll regulate down a 7.4V lipo down to 5V (Cb1, at your suggestion I'll consider a more sophisticated regulator), push it into VBUS, and call it a day.

Thanks all.

May I suggest that a clear drawing - to include the following (at minimum) - be submitted for Amit's review/blessing:

• the specific & identified LPad
• the external battery & any/all voltage regulation (which usually will appear upon a small, external board)
• the exact LPad header/connector pin designators - upon which you intend to supply your 5V & ground
• positioning of any/all switches or board "shunts/jumpers" - which may serve to steer or direct the "externally supplied" 5V

I "know nothing" of vendor's "battery booster pack" - the 4 items above intend to prevent damage/delay - which may result from unclear understanding/identification  and less than, "great attention to detail."

Your William Gibson receives such (undeserved) "airtime/promotion" here - yet remains (completely) silent in all this...  (how is he relevant?)

Feeding (even) a 3V3 LDO Regulator w/3V3 is NOT (usually) recommended.   (3V3 will not be achieved at the regulator's output - some loss will occur.)   In the LPad's case - any/all circuits seeking (and/or depending upon) 5V may fail to meet specification...

Engineering is rarely, "straightforward" almost all serious design consists of the careful, "balancing & trading" of different - often even competing - requirements.   Straightforward (usually) indicates that the objective is trivial or (more likely) not fully/properly understood...

The question essentially was, which pins should be used to provide the Tiva C with an external power supply and at what voltage. This was clear enough. That should be a trivial subject and an (a)-(b)-(c) answer should suffice, possibly with some added tips on the precautions he should have in mind. To my opinion the same applies with most questions asked by people that are seemingly new/inexperienced with a certain board and do not know basic features of it.

In spite of all the "careful" and "balanced" comments, it is a fact that in the end it was the author's colleague that provided the answer needed.

No real need to argue, all. Agree with both sides. Yes, it was a simple question that got a bit out of hand. Also yes, there are many things to keep in mind when working with embedded systems. In conclusion though, I found this to be correct enough:

VBUS is connected to an internal LDO (I had to look this up, but it stands for low dropout) voltage regulator, the TPS73633DRB. The DRB takes an input from 3.5V to 5.5V and outputs 3.3V. Can you power VBUS from a 3.7 LIPO? Yes. But it is a questionable decision because you will soon drop below the 3.5V which is the minimum full-rated performance input voltage to the DRB.

Two alternatives remain. Find a 5-5.5V battery and hook that directly to VBUS. This is tricky because no standard LIPO, NiCd, NiMh, or even SLA batteries come in this range. This leaves us with our final alternative and probably best option: regulate a higher voltage battery down to 5V using an external regulator, and then feed that voltage into VBUS. Two-cell LIPOs (7.4V), NiCd and NiMh (7.2V), and 6V SLA batteries all fall into this range. I'll personally use a 7.4V LIPO and a basic L7805 Voltage Regulator, because my circuit is drawing little current.

I did indeed end up regulating a 7.4V LiPo down to 5V and feeding that into VBUS. Everything worked fine over many uses and still is.

Taking myself off notifications of this thread ;)

Probably you did not understand that I am not the same person as the one that posted the original question. So it was not that I waited 5 months to... squawk(???). I was just looking into the same issue and this was the most relevant thread covering it (unfortunately). In over 20 comments, the only ones making sense came from the conclusions made by the author of the subject. Which was very surprising, given that powering the Tiva C using the power pins is more about instructions rather than engineering.  If someone looks up the same subject for the Arduino board (where more or less you do the same things, only the input voltages differ), he will have the answer in 5-10 lines.

Concluding, I would have to say that if there is something that is harldy forum-material in this thread, that would have to be your writing tone and your (irrelevant and non-technical) remarks.