Removing SMPS on 22" HP monitor(le2201w)

[phorvati]

Hi I am wondering if its possible to replace SMSP on the power board with something that can run off of 12V. I live off of renewable energy and any 120V AC appliance involves using a highly inefficient 12VDC to 120VAC inverter. I would like to bypass this and run the panel off of 12V deep cycle battery and add the 12V/5V buck switcher to power the logic and uC.

Pic of power board is attached.
two threads for LCDs with very similar boards and exact same IC components are
https://www.badcaps.net/forum/showthread.php?t=17348
https://www.badcaps.net/forum/showthread.php?t=31942

Here is what i've tried so far.
1. When AC power cord is plugged in I have 19VDC and 5Vdc on the isolated (logic) side. 19Vdc is the node that powers 0Z9933 ccfl inverter and the two Nfet/-Pfet Ics that feed primary of CCLF transformer. 5V is for logic and uC.


2. When the on-off button is pressed, the 19V node goes to 17VDC(I assume it gets loaded), 5VDC remains 5. The enable pin on oz9933 goes high via the on/off pin on the connector to signal board going high. based on this I assumed that so long as logic gets its 5V and 0Z9933 gets its 19V I should be able to fire it up with on/off button.

3. I got a replacement power board on ebay. Made sure it works by using it in a monitor.

4. Removed the transformer that bridges Mains side to isolated side. Hard wired 12VDC and 5VDC to appropriate nodes. Tried the on/off button but CCFLs did not light up. Enable did go high when on/off button was pressed. I had a Current probe on both DC supply wires and no significant current was detected when on/off button was pressed.

oz9933 datasheet is flimsy but i did find somewhere that it works with wide input voltage range and unregulated DC(but that's more like a marketing writeup as opposed to a good datasheet spec).
I also tried giving the node permanent 19VDC(off of old DELL laptop brick) instead of 12 with the same observation. I did not have 17V DC available.

At this point i am wondering if there is start up procedure that uC looks for during startup and maybe fails to enable ccfl inverter and if there is more to it than just enable going high. Not quite sure what the deal is with the opto going back to SMPS controller and if that's relevant.

Any input would be greatly appreciated.

[budm]

Re: Removing SMPS on 22" HP monitor(le2201w)

Can you show the pictures of the exact points you have connected the power supplies to on the board?
Did the monitor work using the AC before the mod?

[stj]

Re: Removing SMPS on 22" HP monitor(le2201w)

if it works o.k. other than the backlight, i'd say put an led kit in it to halve the current as well as use 12v

this is the link, but the seller is away for china's national holiday.
http://stores.ebay.com/lcd-led-parts

[phorvati]

Re: Removing SMPS on 22" HP monitor(le2201w)

Yellow wire was connected to 5VDC, the red one to 12V and than I tried with 19VDC.
The monitor worked before the mod and its fine in terms of each individual component. There are no bad components or anything like that, I am just trying to get rid of dependency on 120V AC and run it off of 5V and ideally 12V DC source.

As far as the LED I thought about it. Few years ago I tried the LED conversion with another monitor, but I rushed and ended up cracking the glass.

[budm]

Re: Removing SMPS on 22" HP monitor(le2201w)

Where is the power supply return wire/s? It should be on C706 Negative leg.

[phorvati]

Re: Removing SMPS on 22" HP monitor(le2201w)

I took it off of the three transformer ground pin pads see black shaded marking. I also circled part of the ground plane. Ground is shared by both power planes.

Now that i thought about it some I need to verify the following. When transformer is in, and monitor operates off of 120 AC should I be seeing full DC voltage on the plane where red wire is even before the monitor is turned on, i.e AC cord plugged in but enable line low. This is the plane that I am hoping can work with 12V instead of the 17VDC.
I thought I measured 19 VDC on it before "on" button was pressed, but that could be due to caps not bled yet from previous time monitor was on.

[phorvati]

Re: Removing SMPS on 22" HP monitor(le2201w)

So i measured once again and its not a bleed off thing. The node with red wire definitely goes to 19V with respect to ground plane/chassis as soon as AC cord is plugged in. The confusing thing to me is at the same time, I measure 1.2V AC(RMS) with respect to chassis just before the rectifier diode D701 (large pad)? When CCFLS are "on" there is 25V AC RMS there with respect to chassis/GND.
I don't quite understand how this transformer secondary works. How can a 1.2V AC RMS get rectified into 19V DC? I can see how 25V AC RMS gets rectified and loaded down to 17V DC when CCFLS are on.

[budm]

Re: Removing SMPS on 22" HP monitor(le2201w)

The power supply is an always on power supply, it runs 24/7 as long as the power cord is connected to an AC outlet.
You cannot measure the AC Voltage on the secondary side because it is switching power supply running at 70~100KHz, your AC meter bandwidth cannot measure AC at that high frequency. If you are getting 19VDC (drops down to 17V, which is normal) & 5VDC then the power supply is working.
What is the current rating of the external power supply you are using to try to power up the board?

[phorvati]

Re: Removing SMPS on 22" HP monitor(le2201w)

thank you, that explains a lot. I tried with a 19.5V brick from older DELL Inspiron circa 2006. They are i believe at least 3A at 19.5V. 5V came from a PC supply with one of these breakout boards. http://dangerousprototypes.com/docs/ATX_Breakout_Board
The grounds from each DC supply were joined together at the chassis of LCD.

[phorvati]

Re: Removing SMPS on 22" HP monitor(le2201w)

Just wanted to share my findings.
Using 19VDC laptop brick and a 5VDC from ATX breakout board worked. CCFLs light up and I get an image. Using 12V in place of 19V DC does not work. I do get the 12V at all appropriate places on CCFL inverter(0Z9933 ) but CCFLs do not light up. I get 2 second to black syndrome. However no components are damaged. So it appears that 0Z9933 might not work with 12V or perhaps that peripheral circuitry like the sense pin might not be biased correctly for 12V. Perhaps the zeners or coupling cap that sets up the sense voltage. I don't quite know.
If anyone has any information on 0Z9933 beyond that crappy datasheet that would be great.
For now, i believe the way to light the CCFLS with a 12V supply might be to just bypass the 0Z9933 circuit and just get an appropriate Inverter board from ebay.