OM6211U 데이터 시트보기 (PDF) - Philips Electronics

부품명

상세내역

제조사

OM6211U

48×84 dot matrix LCD driver

Philips Electronics 

Philips Semiconductors

48 × 84 dot matrix LCD driver

Product specification

OM6211

22 APPLICATION NOTES

When reading the VM bit in the OM6211 two problems

have been observed: corrupted format and VM bit

toggling.

22.1 Corrupted format

The read-out of the VM bit has a special format, 11110000

for VM = 1 and 00001111 for VM = 0. However,

sometimes a wrong format of the read-out byte can be

observed; the first or the fifth or the eighth bit appears to

be wrong. There are two reasons for this behaviour. When

the first bit happens to be read out at the end of a frame

then it is possible that the first bit belongs to the old VM

value and the 7 following bits belong to the new VM value.

Such behaviour is possible for the first bit only. The second

reason is the violation of the OM6211 timing, if the timing

parameters t2 and t3 (see Fig.17) are violated, then it

results in reading a wrong value for the first, the fifth or the

eighth bit. Thus, to prevent any problems with the wrong

format of the read-out byte, these bits should always be

ignored.

22.2 VM bit toggling

Under certain conditions it can happen that the result of

reading VM is 0 even if the generated VLCD voltage is

correct (VM bit toggles). It is therefore recommended to

repeat the VM read command several times according to

the algorithm described below. This algorithm is based on

the observation that a single reading of VM = 1 (after

numerous readings of VM = 0) is enough to ensure that

the charge pump operation is correct. One possible

method which gives minimum measurement duration is

shown in Fig.31 and described in detail below:

• Perform initialization with Enable OTP and set the

operational parameters (VPR = 159, S = 10, BS = 101,

TC = 1, E = 0 and MY = 0) this results in a slightly higher

VLCD voltage than for normal operation (VLCD = 8.732 V

at Tamb = 27 °C)

• Select DAL = 1 and DON = 1 (for all pixels on)

• After setting HVE = 1 start a loop of a continuous VM

reading (for example, every 1 ms), at first occurrence of

reading VM = 1 interrupt the loop and accept VM = 1

• When the reading is always VM = 0, stop the loop after

a certain time and accept VM = 0. This loop time limit

should be chosen sufficiently long, e.g. 85 ms. Given

that the uncertainty is much less then 0.1%, much less

then 1 ppm is expected to be read out wrong.

For the loop time limit a value of not less than 85 ms is

suggested. It should be noted that the value of 45 ms

specified in Section 11.2.2 means that after at least 45 ms

the VM measurement is possible. In practice it can be

expected that VM is valid earlier than 45 ms. Therefore the

proposed algorithm results in an optimization of the 45 ms

wait time needed to charge the external VLCD capacitor.

So the selected loop time limit of 85 ms consists of 45 ms

wait time and an additional 40 ms of measurement time.

The loop time limit of 85 ms will ensure that even if the first

VM = 1 value for any reason should be missed, there is

always the possibility to hit the next VM = 1 value (note

that the internal VM measurement is made once per

12.5 ms). However, the expectation is that the average

running time of the loop will be less than 45 ms.

There is another possibility for optimization: during the wait

time of the loop (1 ms) other tasks can be performed.

Furthermore the first part of the 45 ms wait time, just after

setting HVE = 1, may also be used for other tasks. For

instance when the first 20 ms are reserved for those tasks,

then the corresponding loop time limit would be 65 ms and

the expected average loop time would be less than 25 ms.

After the VM test is completed the VPR can be set to the

desired value (e.g. VPR = 137) without the charge pump

being switched off.

2002 Jan 17

42

Share Link: