Converting 120v TV to 12v using buck/boost converters

I know this isn't exactly abattery topic but the goal here is to save battery power by not using an inverter in my RV. I picked up an "Insignia NS-24DF310NA19 24-Inch 720p HD Smart LED TV- Fire TV Edition" that has an internal power supply. The input is 120v at0.32 amps AC. So this would potentially be around 12v and 3.2 amps DC. I've attached a picture of the power supply with some comments.

I cracked open the case and found the power supply converts the 120v AC to DC for all outputs. Itsends16.8v to the LCD panel and 12.8v to the mainboard. It appears, if I understand correctly, the mainboard feeds back 5v to the power supply when the TV is on which triggers the 16.8v for the panel.

My thoughts are to use a buck/boost converter to stabilize the +12V connection get 16.8v from the roughly 12v to 14v my battery will provide. I "think" I found one that also has a 5v on/off switch that seemingly would be perfect for turning the 16.8v output on. (amazon.com/dp/B017SM1O98) That is if I understand the "en" port correctly. This buck/boost seems like overkill at 15 amps but it's the only one I can find with the 5v "en" port. Snagged one off ebay for $5.

For the continuous 12.8v I would use another cheaper buck/boost converter that is always on. Unless multiple variable output buck/boost converters are readily available but I couldn't find one.

To avoid parasitic drain I was thinking about adding a switch between the converters and the battery.

• For those more electrically inclined does this all sound correct?
• Any idea what the 4.9v dim pin is for? (See picture)

Important EDIT: I struck out some of my bad readings. The correct readings are in posts #13 and #15.

On the power board, you should probably check the voltage at the transformer output on the low voltage side. in the red/yellow circle below:





If the output is 12V (or there abouts) then you should be able to disconnect the transformer and power the board directly from there.

However, there is anopto ic just to the right of that. Not sure which direction it is facing. Does it trigger the high voltage side, or the low voltage side.

I wouldn't attempt to measure voltages at the transformer, they will be high frequency AC & most multi-meters won't cope with this.
Instead measure at the 3 points in yellow circles this will be the DC on the filter capacitors.





It sounds like you have three easy rails to make, +5, +12, & +16.8.
The "dim" pin is probably control, eg remote control flashes "on" signal to TV, power supply turns on main rails, TV works....
eg +5V might be "always on" for remote to work.

Ahhh yeah, that's right. High frequency. :face palm: I forgot about the frequency issue. Yes, I agree with Readpacket, after the filters.

I wondered if I could tap into the board but had no idea where to start. Are you saying it's likely transformed to 12v DC and later boosted to 16v for the panel?

Before reading the new replies I checked for 12v on the 5 prongs below the transformer but none had a reading. When reading the black chip to the right (I believe it's the opto ic) I read 11.2v on the left bottom prong and 10.2v on the right bottom prong. I measured the prongs on the 2 black chips below the transformer and found that the middle prong on the left chip had the roughly 12.8v. I'll check the black chip circled mid right later.

Trying to tap into the board may be a little more than I understand. I'll probably burn the board out somehow.

EDIT: Struck out some unneeded steps.

Because of the high frequency of the switching, a standard DMM won't get a voltage reading, or a reliable one, on the windings on the transformer. That's why Redpacket suggested the other chips to test from.

It's possible it's converted from 120VAC that is then rectified, then inputted into the transformer as 120VDC pulsed which then outputs as 16VDC pulsed, then smoothed out by the filter capacitors that Redpacket pointed out. (I'm guessing on the 16V output). Redpacket is the switching psu guru

Korishan said:

It's possible it's converted from 120VAC that is then rectified, then inputted into the transformer as 120VDC pulsed which then outputs as 16VDC pulsed, then smoothed out by the filter capacitors that Redpacket pointed out. (I'm guessing on the 16V output). Redpacket is the switching psu guru

Click to expand...

Yes, that's how it works.
Above the transformer there's a transistor on a small aluminium heatsink - this is switching the approx 160VDC (rectified 120VAC) into the transformer.
The transformer will have 3 windings,
- one makes the +5V rail
- one makes the +12V rail
- one makes the +16V rail
See the points I marked for you- those devices are all types of diodes.

Be careful with any part inside the hashed border area (left side mostly) as this has lethal voltages & if shorted, the flash/bang will be nasty.Also they don't become safe until several minutes after power off.

I'm a little confused.

With the TV off, and using the case as the ground, the +5V is unpowered but the +12V and +16V arepowered. If I use the ground wires, instead of the case, none are powered with the TV off.

Is the +5V coming from a rail on the power supply or is it coming from the mainboard?(I attached apicture of themainboard) This makes me believe that once the TV is turned on the ground wires are grounded to the casewhich passes +12V to the mainboard that then bucks and sends back +5V to the power supply?

If this is true, what purpose does sending +5V back to the power supply accomplish?I'm also confused because some power must always be on in order for the TV to receive contact from the remote control.

EDIT: Struck everything out as my readings were bad. I removed the power supply preventing it from grounding to the case. My later readings in post #13 and #15 are accurate.

So I know that this isn't the answer you want - but really the way to do it is to look for a TV with an external power supply - as in, they pipe DC straight into it.

I have two LCD TVs that I have that I run on DC that I converted. The easiest is to just use an external power supply (BTW, TVs are so cheap now, it's almost worth it to just go and get one) One had an external power supply, and the other took in straight 120VAC, rectified it and bumped it down to 19VDC. I believe it had a transformer too to get to 5v and 3v for other goodies -. I tied into the 19vdc rail with a buck converter from my 12v supply and somehow it all magically works. It really shouldn't, but it works great. However, I think that's a fluke more than anything. Like how it somehow works its way down to the other voltages was a miracle more than it was intelligent decision (I have no idea what I'm doing) so for those out there interested, the easiest and safest way is to just aim for a TV with an external power supply.

MilkMan said:

I'm a little confused.

With the TV off, and using the case as the ground, the +5V is unpowered but the +12V and +16V arepowered. If I use the ground wires, instead of the case, none are powered with the TV off.

Is the +5V coming from a rail on the power supply or is it coming from the mainboard?(I attached apicture of themainboard) This makes me believe that once the TV is turned on the ground wires are grounded to the casewhich passes +12V to the mainboard that then bucks and sends back +5V to the power supply?


If this is true, what purpose does sending +5V back to the power supply accomplish?I'm also confused because some power must always be on in order for the TV to receive contact from the remote control.

Click to expand...

So your multi-meter black probe should be on one ofthe pins you've marked as"grounds connected when TV is on" ie the orange or green wire,this is the power supply's 0Vrail.
With the red probe you're testing the 3x circled points I added to the pic.

The case of the TV might not be connected to 0Vor might be switched by a part of the circuit.

Mikethezipper has a good point about picking a device with a single voltage external PSU.

Thank you for the help. I'll reply once I've had some more time to tinker and test everything or I run into a problem.

I picked this particular TV, despite not having an external power supply, because it's a Fire TV. One remote is nice, less power usage without an extra media box, and the Fire OS is android based so I can side load Kodi to watch my content without internet. I was planning to plug a portable hard drive into the back of the TV but that has its own problems. Fire OS only reading fat32, difficulty formatting 4TB to fat32, and not enough power from the USB port. So now I'll likely have a Raspberry Pi or travel router with samba to read the hard drive as exfat. I'll bring in power from the 12v line instead. So with a Pi I could have grabbed a dumb TV after all. I hope someone discovers a path to rooting this TV one day as well.

But I enjoy the project either way and I've learned a lot already.

A while back I had a samsung 32" tv that was a smart tv- I believe it had an external supply although with the samsung premium it was like $220 or so. I ended up returning it because when I measured the power draw it was waaay to high. The "smart" feature on a couple tvs I checked added quite a bit of power draw, even when I wasn't doing anything "smart" with it. I went with a simple panel - and a plug-in chromecast. So sometimes I use that, and sometimes I just hookup my computer via hdmi. With the computer, I end up using more power than the smart TV used, but I get more funtionality out of it. Definitely let us know how it goes since it'll probably help when I change RVs later and need to get a new TV.

Mikethezipper said:

A while back I had a samsung 32" tv that was a smart tv- I believe it had an external supply although with the samsung premium it was like $220 or so. I ended up returning it because when I measured the power draw it was waaay to high. The "smart" feature on a couple tvs I checked added quite a bit of power draw, even when I wasn't doing anything "smart" with it. I went with a simple panel - and a plug-in chromecast. So sometimes I use that, and sometimes I just hookup my computer via hdmi. With the computer, I end up using more power than the smart TV used, but I get more funtionality out of it. Definitely let us know how it goes since it'll probably help when I change RVs later and need to get a new TV.

Click to expand...

That's a good point. I'll be sure to measure the true power usage as I go. I estimated at most 3 amps at 12v.

Redpacket said:

I wouldn't attempt to measure voltages at the transformer, they will be high frequency AC & most multi-meters won't cope with this.
Instead measure at the 3 points in yellow circles this will be the DC on the filter capacitors.





It sounds like you have three easy rails to make, +5, +12, & +16.8.
The "dim" pin is probably control, eg remote control flashes "on" signal to TV, power supply turns on main rails, TV works....
eg +5V might be "always on" for remote to work.

Click to expand...

Redpacket said:

So your multi-meter black probe should be on one ofthe pins you've marked as"grounds connected when TV is on" ie the orange or green wire,this is the power supply's 0Vrail.
With the red probe you're testing the 3x circled points I added to the pic.

The case of the TV might not be connected to 0Vor might be switched by a part of the circuit.

Mikethezipper has a good point about picking a device with a single voltage external PSU.

Click to expand...

I realized that many of my earlier measurements mayhave been inaccurate because I physically removed the power supply which prevented it from grounding to the case. (I'm a dummy)This changed many of my readings.

I measured the voltages at all 3 points you listed. I used the grounds from the orange and green wire. (Pins 5 and 7 though pins 6 and 8 are also grounds)I tested with the TV on and the TV off.

TV on
Mid right circle = +11.5V (Reads high +12V/+13V butsettles mid +11V)
Bot left circle = +12.86V (Not much flux)
Bot right circle = +9.65V (Not much flux)

TV off
Mid right circle= +16.7V
Bot left circle= +12.72V
Bot right circle= +12.2V

Other readings
The +16.8V reading to the LED panel (top connector)I took earliernow reads +13.36V with the TV on. (Using grounds within the top connector) Pin 1 (dim, brown wire) of the connector at the bottom right reads +5.1V but pin 2 (on/off, black wire) reads +2.5V with the TV on. (Using pin 5 ground,orange wire)These all read a hair over +0V with the TV off.

Pins 9, 10, 11, 12and 14 of the bottom right connectorall read +12.7V with the TV on or off. So the mainboard receivespower whether the TV is on or off.

• Do you still believe there are 3 rails? (+5V, +12V, and +16V)

Good job with the measurements there.
Yes there's 3 rails. +5V, +12V, and +16V & the 12V appears to have two halves: one for audio & one for other stuff (see square white box you've labeled "pin break down").

I put +12V's directly to the middle prong of each +12V diode. This seemed to power everything normally. All of the readings mostly matched prior readings. So I assume there's some buck/boosting going on. I'll be purchasing a buck/boost to keep a stable +12V

• Booting the TV up used between 1 amp and 1.6 amps at +12V
• Idling at the main menu used about 1 amp
• While watching an over the air channel with moderate speaker volumeit used 1.6 amps (Did not test streaming content)
• While charging my phone inthe USB port at +5V 500 ma and watching an OTA channel it used 1.9 amps at +12V.

I'm impressed with how little power it uses.

EDIT: I added a picture to show how I quickly tested sending +12V to the 2 diodes. I started with only +12V on the left diode but the screen wouldn't power on. Later I'll cleanly solder and protect the connections.

MilkMan said:

I put +12V's directly to the middle prong of each +12V diode. This seemed to power everything normally. All of the readings mostly matched prior readings. So I assume there's some buck/boosting going on.

Click to expand...

That sounds like a good outcome
Yes you're right about the buck/boost - you can see 2x inductors that are likely being used for this, one is just above the multi-pin connector (middle right)& one in bottom right corner.
The back of the board also has at least 2x switching controller chips on it.

Nice it doesn't use much power too!

MilkMan said:

I'm impressed with how little power it uses.

Click to expand...

Now, have you tested how much wattage it uses running on AC? It would be interesting to see what the difference is between AC power and DC power.

MilkMan said:

I put +12V's directly to the middle prong of each +12V diode. This seemed to power everything normally. All of the readings mostly matched prior readings. So I assume there's some buck/boosting going on. I'll be purchasing a buck/boost to keep a stable +12V

• Booting the TV up used between 1 amp and 1.6 amps at +12V
• Idling at the main menu used about 1 amp
• While watching an over the air channel with moderate speaker volumeit used 1.6 amps (Did not test streaming content)
• While charging my phone inthe USB port at +5V 500 ma it used 1.9 amps at +12V.

I'm impressed with how little power it uses.

Click to expand...

Awesome man! Glad to hear it worked for you without having to put in separate rails for each voltage.
You mind editing your initial post to put in the exact model you used so maybe someone else can follow that road later on?
12W is pretty good considering it's a smart TV too - What size is it? I went with an LED tv because of their lower power use, but it isn't "smart" and it uses that same amount. Mines a 32" samsung - but I use an external 19v boost converter on mine (I chose it for the external power supply and size)

Redpacket said:

That sounds like a good outcome
Yes you're right about the buck/boost - you can see 2x inductors that are likely being used for this, one is just above the multi-pin connector (middle right)& one in bottom right corner.
The back of the board also has at least 2x switching controller chips on it.

Nice it doesn't use much power too!

Click to expand...

Thank you for the help everyone. Lots of learning for me and I appreciate the help and guidance.

Korishan said:

Now, have you tested how much wattage it uses running on AC? It would be interesting to see what the difference is between AC power and DC power.

Click to expand...

I would like to know but I don't have a meter to read AC current. I'll keep it in mind when I do eventually purchase one.

Mikethezipper said:

Awesome man! Glad to hear it worked for you without having to put in separate rails for each voltage.
You mind editing your initial post to put in the exact model you used so maybe someone else can follow that road later on?
12W is pretty good considering it's a smart TV too - What size is it? I went with an LED tv because of their lower power use, but it isn't "smart" and it uses that same amount. Mines a 32" samsung - but I use an external 19v boost converter on mine (I chose it for the external power supply and size)

Click to expand...

Mine is actually an LED TV too. I cleaned up some of my misleading posts by striking out comments. I also added a picture of my "test" connections in post #15. I'll do a little more cleaning and add pictures when I receive the converters. The converters will burn some extra power and the Pi/HDD will use some too.

While I'm at it I might as well list the converters I purchased. If something does not work I'll strike it through and edit later. The more patient you areto wait on shipping from Chinathe lower the prices these are on ebay. Quality can be an issue with these though.
75W DC Buck Boost Voltage Converter Constant Current Module Step Power Up- $4.95
DC Converter Buck Module 12V Convert To 5V 5V 3A Usb Output Power Adapt 15W DC- $5.72 (Power for the Raspberry Pi and portable hard drive)
https://www.ebay.com/itm/60W-DC-Buc...e=STRK:MEBIDX:IT&_trksid=p2060353.m2749.l2649 - Previous converters did not work. See post #22.
Low-profile Inline Controller Dimmer Switch for LED Strip Light, Bare Wire- $3.45 ... maybe I should have just purchased a switch froma local store... still may but good up to 12 amps.
In Line Standard Blade Fuse Holder Splash Proof for 12V 30A Fuses Car Bike 5pcs- $0.99

If I manage to root this TV one day I can install apps on the TVto read exfat or ntfs partitions and instead connect the drive directly to the TV. Though I imagine I'll have some fun with the Pi connected too. It has RetroPie currently.

The exact TV I purchased was the "Insignia NS-24DF310NA19 24-Inch 720p HD Smart LED TV- Fire TV Edition". I picked it up for $80 after a 20% Google Express discount. It's also on Amazon and Best Buy's website. (Best Buy SKU 6247347) 1080P or 4K would have been better but my focus was more so on the smart Fire TV (Android based) aspect and 1080P starts at 39". The power usage would havebeen higher too.

https://express.google.com/u/0/prod...0CAAQ0FUoPWoYChMItc76jtLY4AIVBQgBCh1ZqgS6EOYH
https://www.amazon.com/Insignia-NS-24DF310NA19-24-inch-720p-Smart/dp/B07FPRZ69X
https://www.bestbuy.com/site/insign...-hdtv-fire-tv-edition/6247347.p?skuId=6247347

Man, you're awesome. Thanks for taking the time to document all this.
Something that I'm wondering, is how efficient all these converters really are. I am using a few myself, and never really gave it much thought since the advertisement says "Over-unity, greater than 100% efficiency" but I've had to design a couple buck converters myself over the past couple months and I'm seeing that's a load of BS. Basically, all buck converters have a peak efficiency at only certain settings. Like a 75W converter will only work at >90% efficiency if the current draw is within a certain window. Out of that window, it drops like a rock. For example - I have a buck converter where if you look at the datasheet for the actual chip, it'll do up to like 90% efficiency when you use it in the 70-90% of it's rated power output. If you have a 75W converter, and you are only using 12W, it's very possible you could be in the 70% efficiency range or lower. That's why they say what their peak efficiency is. It's a horrific truth I stumbled upon after many knowledgeable people on this forum told me but didn't understand .. or want to believe, until I saw the actual datasheet. Many datasheets I've seen, the efficiency curve drops exponentially completely off the graph once you get away from it's designed power output - so I have no idea what sort of efficiency they are really using. And since it drops off the graph, the only way we'd know is by measuring it empirically. I bought a 200W converter since that's my peak load on my TV/Sound system - but my usual load is probably closer to 30-40Watts combined. I really should check that..

Also, when I look at these buck converters from china, basically none of them are using the recommended capacitors that the actual buck chip mfg recommends. Since the real capacitors are large and expensive, they just grab from a box of whatever is the cheapest thing and just slap them on there. Really - I don't think that affects the efficiency, but it just means the system will be noisier and have more fluctuations than it probably would normally have. Maybe someone who actually understands electronics can chime in here since I'm just a crazy person who found a soap box on a corner and decided to stout spouting stuff about the end times.

Ah, yeah 24" would explain why my tv uses more power since it's a 32". I'll say what though, for supposeddly being a 1080p tv from samsung, 1080p content looks like garbage on this thing. I don't think you missed out on much I have a feeling like computer monitor resolutions are much more trustworthy than what TVs are advertised at. And at a lower resolution, there's less thinking the brains of the system has to do, so also less power use.

The irony is I spent soo much time looking for a 32" 1080p TV - only to find out that my hotstop auto-throttles all digital media to 720p max