Question on powering LCD backlight

[bigal_scorpio]

Hi again,

Just had one question answered and now found another!

Just realised that my LCD display I was planning on using has not got the standard LED backlight, but instead has an Electro Luminescent pad.

I know that they require an AC voltage to drive them but can't find anywhere that has info on building a simple circuit to drive it! One that wouldn't kill it anyway!

So if anyone has a simple way of doing this I would be very grateful, I thought about a 555 timer circuit but it seems to be the only 555 circuit that I CAN'T find on google

Thanks again for looking...........Al

 

[Torben]

Sounds like you need an inverter to get the high AC you need. I would suggest just buying one, but you don't say what the make and model or specs of the LCD display are so it's hard to say any more than that.

Try Googling for "electroluminescent backlight inverter schematic"--I found **broken link removed** quickly, but I'm not sure it is suitable for your backlight for two reasons: 1) I don't know what your module is; and 2) I am no inverter expert.


Good luck!

Torben

 

[bigal_scorpio]

Can't source the ICs

Sounds like you need an inverter to get the high AC you need. I would suggest just buying one, but you don't say what the make and model or specs of the LCD display are so it's hard to say any more than that.

Try Googling for "electroluminescent backlight inverter schematic"--I found **broken link removed** quickly, but I'm not sure it is suitable for your backlight for two reasons: 1) I don't know what your module is; and 2) I am no inverter expert.


Good luck!

Torben

Thanks Torben,

I followed the link you sent and indeed found a schematic, but as for making one it seems that all the schematics available seem to show special EL ICs that I can't find anywhere to buy.

Thats one reason I was thinking I may be able to rig something up, maybe involving a 555 running in astable mode with its output amplified through a transistor or two and maybe a small audio transformer?

Sadly like you I am no expert on EL (well I'm no expert on anything really ).
So if anyone has any ideas no matter how strange, or if any experts fancy helping me design something................

Any thoughts?

Al

 

[bigal_scorpio]

The specs of the LCD

BTW I forgot to add that the LCD is a 2 x 20 display that has the marking
LM032LN believed Hitachi

Thanks for looking..........Al

 

[Torben]

Hi Al,

Check this thread for some more information and a link to an appnote which might help.

Personally I don't see why you couldn't use a 555 at 400Hz to do this, except you'd need to use it to drive a transistor which can handle the high voltage. The circuit given in Figure 2 at **broken link removed** uses a Zetex FMMt458 as the final drive transistor. Are you cool with SMD devices?

I could just be making this more complicated than it needs to be.


Torben

 

[bigal_scorpio]

Further developements

Hi Al,

Check this thread for some more information and a link to an appnote which might help.

Personally I don't see why you couldn't use a 555 at 400Hz to do this, except you'd need to use it to drive a transistor which can handle the high voltage. The circuit given in Figure 2 at **broken link removed** uses a Zetex FMMt458 as the final drive transistor. Are you cool with SMD devices?

I could just be making this more complicated than it needs to be.


Torben

Hi again Torben,

I like the look of that circuit, and I don't mind using the odd SMD component, but alas I dont have any of the ones in the circuit.

I do however have what I hope would be an equivalent for both the FMMT458 and the FZT857, its the JE340p, sheet attached!

I also have a good few ICM7555 ICs, sheet attached,was thinking would be a possible substitute for both SCT1555 ICs, what do you think?

I am also trying to keep to parts I already have as I will need to make 4 of these inverters eventually to use the 4 LCDs I bought (next time I will make sure they are LED backlights) Doh!

Anyway thanks for the help so far, its much appreciated........Al

 

[Torben]

Hi again Torben,

I like the look of that circuit, and I don't mind using the odd SMD component, but alas I dont have any of the ones in the circuit.

I do however have what I hope would be an equivalent for both the FMMT458 and the FZT857, its the JE340p, sheet attached!

I also have a good few ICM7555 ICs, sheet attached,was thinking would be a possible substitute for both SCT1555 ICs, what do you think?

Anyway thanks for the help so far, its much appreciated........Al

Well, there might be something in the datasheet for the transistor I'm not seeing which means it wouldn't work, but it seems to me to meet the requirements as far as being able to handle the amperage, frequency, and voltage needed. Same for the ICM7555. If it were me, I'd try it.


Good luck!

Torben

[Edit: But I'd be sure to check the circuit's output before hooking up the first backlight.]

 

[bigal_scorpio]

Built it!

Hi Torben,

Built the inverter circuit this afternoon! Then spent 4 hours trying to get it to work!

Could be I got something wrong in my translation to Vero board, or maybe the circuit isn't as tolerant of the parts that I had to substitute?

Anyway I am going to have a go at breadboarding it instead I think or do you think it would be worth a try on Proteus or Multisim, just to check the original circuit's operation?

BTW when I did measure the output with my DMM there is about 9v on AC, but is there any possibility that my DMM is not measuring its output correctly?
I was wondering if the frequency was way out would it give a false reading on the meter?

Any thoughts welcome.............Al

 

[audioguru]

A DMM is accurate at 50Hz and 60Hz. The inverter operates at a much higher frequency so a DMM reads the AC voltage low. An ocilloscope would be accurate.

 

[Torben]

Hi Torben,

Built the inverter circuit this afternoon! Then spent 4 hours trying to get it to work!

Oy. I know what that feels like.

Could be I got something wrong in my translation to Vero board, or maybe the circuit isn't as tolerant of the parts that I had to substitute?

Could be. And could be.

Anyway I am going to have a go at breadboarding it instead I think or do you think it would be worth a try on Proteus or Multisim, just to check the original circuit's operation?

I'd definitely sim it before I built it if I could (I use LTSpice myself). Then I breadboard, then build. I'd skip the breadboarding on RF and power circuits but for this I'd think it would be fine as long as the board is in OK condition.

BTW when I did measure the output with my DMM there is about 9v on AC, but is there any possibility that my DMM is not measuring its output correctly?
I was wondering if the frequency was way out would it give a false reading on the meter?

Any thoughts welcome.............Al

See audioguru's note. If I get a chance tonight I'll take a shot at simming it in LTSpice and see what I get. I'll check the output against the datasheet and see it it's putting out a sane voltage/frequency for the device. Of course a sim won't tell you for sure that something will work, but my workshop is currently in "metal shop" mode so putting something together won't be possible.


Torben

 

[kchriste]

I was wondering if the frequency was way out would it give a false reading on the meter?

Some good quality (Fluke) DMMs are good up to 10-100Khz. A cheap meter may not go as high. Look at the spec sheet for your meter if you have one. If there is no spec sheet, assume it is only accurate up to 500Hz or so.

 

[Torben]

Well, I simmed it. I substituted a couple of things for which I didn't have or couldn't find models: all the diodes are 1N4148s and the transistors are 2N5550s (which have a Vceo of 150V instead of 400V, but it's what I have right now). And the 555s are just idealized NE555 models.

Short story: it works. Sorta. In fact, it shows way too high a voltage, since the datasheet for your unit indicates that it wants ~100V at ~400Hz. This thing is pumping out ~1.6kVDC, with spikes to ~2.9kV at ~42kHz. At least, that's what the sim says.

I'm kinda bushed right now but I'm wanting to tinker with this thing for a while yet. It actually seems to me now that it should be easier to get 100V at 400Hz but it'll take me more reading before I figure it out--I'm a hobbyist, not a switchmode power supply engineer.

At any rate, at least I think this shows that the schematic itself should be OK. I'd have to actually build it to see what it would do in real life (er. . .but you knew that already. . .).

Anyway, that's all I have for now.


Torben

 

[mvs sarma]

Hi again,

Just had one question answered and now found another!

Just realised that my LCD display I was planning on using has not got the standard LED backlight, but instead has an Electro Luminescent pad.

I know that they require an AC voltage to drive them but can't find anywhere that has info on building a simple circuit to drive it! One that wouldn't kill it anyway!

So if anyone has a simple way of doing this I would be very grateful, I thought about a 555 timer circuit but it seems to be the only 555 circuit that I CAN'T find on google

Thanks again for looking...........Al

Hai, Unless you are specific of using this only for reasons known to you, Perhaps you may avoid further investing on this and rather go for conventional LED type backlit and with display pin-out to suite your project. I feel it may not cost that much as a ready made inverter.

As reg using the present one, what voltage level you need for the back lit to work? Perhaps you can use an clock available on your microchip and using a transistor and inductor, you can make a charge pump and i wonder whether such arrangement would power it. You need not rectify the output.

 

[Torben]

Hai, Unless you are specific of using this only for reasons known to you, Perhaps you may avoid further investing on this and rather go for conventional LED type backlit and with display pin-out to suite your project. I feel it may not cost that much as a ready made inverter.

As reg using the present one, what voltage level you need for the back lit to work? Perhaps you can use an clock available on your microchip and using a transistor and inductor, you can make a charge pump and i wonder whether such arrangement would power it. You need not rectify the output.

That's what I was getting at with the "easier to do" idea. I'm thinking maybe the circuit from the ED site is overkill for a LCD backlight. This backlight only wants ~100V at ~400Hz, according to the datasheet.


Torben

 

[mvs sarma]

That's what I was getting at with the "easier to do" idea. I'm thinking maybe the circuit from the ED site is overkill for a LCD backlight. This backlight only wants ~100V at ~400Hz, according to the datasheet.
Torben

I feel that when a new LCD display with LED back-lit is cheap enough- no point in designing a aux psu for old version LCD. Perhaps if one has an LCD panel without back lit and pin compatible, better substitute and use ad Normal LCD for day light.
Alternately one may have faulty displays where the back-lit is working and it could be salvaged and used for this.

 

[bigal_scorpio]

EL problem further

Hi to everyone,

Thanks for simming the circuit Torben, the infor about the peaks and such is nice to know.

As for Sarma's suggestion about just giving up and buying new LCDs, well I got the displays cheap and because my means are limited I usually have to work with what I have to hand. I know in an ideal world none of us would have to make do, but sadly I am still looking for that utopian bliss

Also Audio Guru suggests that an Oscillioscope would be better to measure the output, but sadly a scope is not yet on my agenda due to limitations of cash, but I live in hope, and considered making one of the simpler PIC based PC scopes? Thoughts?

Anyway I wondered if there was a different way of measuring the output, no doubt someone knows of a fiendish method that doesn't involve a scope?

Further to Torbens suggestion that the circuit is overkill, I was wondering if simply putting the output of a 555 astable circuit running at the required frequency through a small audio output type transformer could be possible, again I am thinking this simply because I have some of them to hand, and from the specs on the backlight it seems that very little actual current would be drawn.

Which brings me to one final question - would I need to limit the current to the EL as is required when using LEDs or is the EL somehow self regulating?

Thanks for all the suggestions and help guys, keep it coming please.

Al

 

[Torben]

Hi to everyone,

Thanks for simming the circuit Torben, the infor about the peaks and such is nice to know.

No problem.

Further to Torbens suggestion that the circuit is overkill, I was wondering if simply putting the output of a 555 astable circuit running at the required frequency through a small audio output type transformer could be possible, again I am thinking this simply because I have some of them to hand, and from the specs on the backlight it seems that very little actual current would be drawn.

It's possible, at least mathematically. There are simple circuits like **broken link removed**. I played around with it for a while in LTSpice and came up with this:

Try whatever transistors you can--I'd start with TIP41 and TIP42; I just used the ones listed because I had models for them.

[Edit: just to clarify: that's not my circuit; I just modified Tony van Roon's to do what I wanted.]

Another note: I find that the sim hits around 400Hz output with the pot's wiper at around 0.75 (turned three-quarters up).

Which brings me to one final question - would I need to limit the current to the EL as is required when using LEDs or is the EL somehow self regulating?

Thanks for all the suggestions and help guys, keep it coming please.

Al

I don't know the answer to that question. The circuit I posted delivers about 20mA into a 5kOhm load, but I don't know what kind of load the EL presents. The datasheet for the EL panel doesn't show any current limiting on the example schematic, but that doesn't mean that the driver chip used doesn't limit the current.


Torben

 

[bigal_scorpio]

Circuit query

Hi Torben,

Just about to start building your circuit when I noticed something iffy. Are C3 and L2 doing anything? There is a link that bypasses them, is this intentional?

I must admit to being a bit lacking in knowledge about inductors so if the bypass is intentional please let me know.

Thanks again.....Al

 

[Torben]

Hi Torben,

Just about to start building your circuit when I noticed something iffy. Are C3 and L2 doing anything? There is a link that bypasses them, is this intentional?

I must admit to being a bit lacking in knowledge about inductors so if the bypass is intentional please let me know.

Thanks again.....Al

The bypass is just to get the darn thing working in LTSpice. Myself, I'd try it just leaving bypass out to start with. I forgot to mention that in my post--sorry. The cap and inductor are just to try to condition the signal a bit so the transformer isn't getting a square wave (which I *think* would result in an impulse wave output).

Hm. I was thinking I'd get a chance to build this later on but tonight but I forgot my wife's baseball game.


Torben


Torben

 

[bigal_scorpio]

Bypass

Hi Torben,

Thanks mate, I'll get building and good luck with the ballgame

BTW just had an idea about testing the output on the other circuit I made, wondered if I could try a Neon lamp (from an old lighted switch) across the output, but I didn't get a flicker!

Al