LCD Contrast Bias Problems

[[email protected]]

I'm using an Okaya RC2004ARS-DSC-A (RC2004D, the baby 20x4) display in
a 3.3V system. I originally planned on using a negative voltage charge
pump driven by a PWM pin on a PIC (a circuit I've used before
successfully.) When I couldn't get that to work I tried a ICL7660 neg
chg pump. No luck on anything. Everything logic-wise appears to be
operating normally, it init's fine, etc. When I put 5V on the system I
can actually get the contrast up enough to start to see the
characters.

There is an Okaya appnote that speaks of internal and external
contrast adjustment on their modules. However, none of the jumpers
they describe appear on the RC2004. So I'm assuming it doesn't have
that capability (which could be causing the problem if somehow the
internal contrast selection was jumpered on.)

When I probe the Vo pin, I get an impedance of about 3K ohms. When I
test the ICL7660 without it connected to the V0 pin I get a good V of
about -3.2V from a 3.3V supply. However, when I connect the ICL7660
output to the Vo pin all I get is -0.7V when scoped. The 4V bias
obviously isn't enough when I need at least 5+ volts to start to see
the characters when tested with a 5V Vdd. BTW, one would expect that
when using the ICL7660 that would produce a bias of about 6.5V and the
screen would be over contrasted and be dark. No such luck, it's as if
there's no bias at all.

Any ideas or suggestions would be hugely appreciated,

John

 

[[email protected]]

OK.

I've narrowed it down the Microchip MCP41010 SPI Digipot. If I replace
the 41010 with a regular old pot everything works normally. I
remembered there was a 1mA limit on the wiper of that part so I added
a 4.7K current limiter. But that doesn't change anything.

Any ideas?

Thanks,

John

 

[Miguel Gimenez]

OK.

I've narrowed it down the Microchip MCP41010 SPI Digipot. If I replace
the 41010 with a regular old pot everything works normally. I
remembered there was a 1mA limit on the wiper of that part so I added
a 4.7K current limiter. But that doesn't change anything.

Any ideas?

Thanks,

John

The Digipot needs all inputs to be between the power rails. Your
negative voltage is being clamped by an internal diode to Gnd.

 

[[email protected]]

OK. That makes sense. Thanks for the info. I think I see where that is
mentioned in the datasheet.

Does anyone know of a digipot that isn't limited in this way?

Thanks,

John

 

[Adrian Jansen]

OK.

I've narrowed it down the Microchip MCP41010 SPI Digipot. If I replace
the 41010 with a regular old pot everything works normally. I
remembered there was a 1mA limit on the wiper of that part so I added
a 4.7K current limiter. But that doesn't change anything.

Any ideas?

Thanks,

John

Sounds like you are creating a negative voltage with the 7660, then
adjusting it with the digipot to feed the LCD.
Do it the other way. Create an adjustable positive voltage, either from
a PWM or otherwise, then feed that to the 7660 to invert it, and feed
the output direct to the LCD.

--
Regards,

Adrian Jansen adrianjansen at internode dot on dot net
Design Engineer J & K Micro Systems
Microcomputer solutions for industrial control
Note reply address is invalid, convert address above to machine form.

 

[[email protected]]

OK. That makes sense also. I had thought of maybe using the 7660 (as a
voltage doubler) to supply the LCD and then use the digipot normally
as a contrast adjust between Vss and Vdd, of course Vdd for the LCD
now would be 2 x supply voltage to the 7660.

Thanks again,

John