| Schematics |
|
updated Monday, August 11, 2003 | email: pstewart+atsign+gwi+dot+net |
MN83803AK timing driver in electronic viewfinder for the PV-D705 camcorder
MN83803AL timing driver in electronic viewfinder for the PV-950B camcorder
IR3y05y video processor in PV-950B camcorder electronic viewfinder, with MN83803AL
Discrepancies within and amongst diagrams & data sheets
Inferences Some conclusions.
Related manuals other camcorder manuals
RGB input conditioner to go with Kristian's design design idea
Lacking application notes for the MN83803AK and schematics for the Cybermaxx 120k video system, it occurred to me that schematics in Panasonic camcorder repair manuals might be of help.
It turns out that these manuals are inexpensive and full of information (though full of typographical errors as well).
My first search, kindly conducted for me by Panasonic's literature search folks, was for camcorders that used the MCL0712A03 LCD (the one this web page is dedicated to), but their database contained no reference to it; my second search was for a camcorder with a 505 x 230 pixel EVF (electronic viewfinder) LCD like ours, figuring that a similarly configured LCD might use the same driver chips as the Cybermaxx. I found a camcorder with this EVF LCD configuration, and thinking this would be sufficient, I ordered the manual for it; it used a highly integrated ASIC combining the functions of video processing otherwise provided by chips like the Sharp IR3y05y with the LCD timing functions the MN83803Ax series implements elsewhere. The LCD it used was the MCL0512B01.
I've since misplaced this manual, so I don't know offhand which camcorder models it is for.
So I obtained a list of camcorders that used the same LCD as this one, the MCL0512B01, from which I ordered another camcorder manual, for the PV-D705 palmcorder series. This series uses the MN83803AK and the AN2523FAP/K video signal processor.
MN83803AK driver chip Click here for a detail of the color EVF schematic diagram from the PV-D705 series repair manual, showing the MN83803AK. This is the diagram Kristian Bognaes used to create his application board for the MN83803AK.
Subsequently I ordered the manual for the PV-940 and PV-950 series camcorders, which uses the IR3y05y for its video signal processor (converting composite video to RGB color information), and the MN83803AK/K for its LCD timing controller. [Is the /K here another typo, that should read /L instead?]
MN83803AL driver chip Interestingly, this manual came with a supplement that shows an MN83803AL timing controller chip-- affording to us the opportunity to assess the claimed pin-compatibility of the -AK and -AL chips prior to soldering them into application boards. (A precaution suggested to me because of the changes in Panasonic's information to this effect, and the grain of salt with which we must take any information coming from a large bureaucracy).
IR3y05y video processing chip Click here for a detail of the EVF schematic from the PV-950-B / PV-950B-K repair manual supplement, showing the application of the IR3y05y (cf. the schematic detail above, from the same diagram).
Sharp Electronics Corporation has provided me with excellent hardcopy documentation on the IR3y05, in a very timely fashion. Their Literature Fulfillment and Technical Support departments have been friendly, helpful, and highly competent. I'm very pleased with them, and without their help, this project would not have a chance of succeeding.
The telephone number for their Literature Fulfillment Department is 1-800-642-0261.
Sharp's FastFax (faxback) service has at least in the past carried the ir3y05 data sheets. Call 1-800-833-9437 (North America only) to request the Sharp FastFax index, where you can find the FastFax document title you will need to enter from your FAX machine or FAX software. (bear in mind that the quality of a FAX document may not measure up to that of the hardcopy the Literature Fulfillment Department will supply-- but it's worth a try. If only the last, large (2-page?) schematic with the Japanese script around the edges looks too fuzzy on your FAX copy, but the other sheets are legible, don't worry: That particular schematic is similarly illegible on the hardcopy, apparently originating there also from a FAX.
If you need the documentation on the IR3y05, cannot find it anywhere, and I have not put it on this web site yet, let me know.
You may also see Sharp's microelectronics web page (http://www.sharpsma.com) for I.C. information.
The data sheets they provide for the IR3y05 look excellent, providing detailed pinouts, detailed descriptions of pin function, block diagram, basic connection diagram, typical application diagrams, Input/output circuit type diagrams (for i/o pins), gamma correction diagrams, FRP / RGB output diagrams (cf. Nate Caine's explanation at Halted Specialties' web site ), precautions, absolute maximum ratings, recommended operating conditions, AC and DC characteristics, input waveform and sync signal diagrams (looks like standard NTSC values, e.g. 0.286 volt sync depth, 0.714 volt video signal ceiling), and various test circuits.
IR3y05y supply Vcc2
From these data sheets, the first item that leapt to my attention was the absolute maximum rating for Vcc2: 14 volts. Panasonic's EVF runs the IR3y05y at 14 volts, according to its schematics: Is this the reason we find the 'y' at the end of the chip's name? Is this a variant of the IR3y05 with elsewhere-documented different DC characteristics?
Correction: A closer look at this schematic reveals test voltage readings of 13.7 volts at this pin on the IR3y05y-- within the IR3y05's rated maximum operational voltage of 13.75volts. The FPC connector between boards in the electronic viewfinder labels this as 14 volts, leading to the confusion. For more about supply voltages, see "Care and Feeding of the Cybermaxx LCD" at the bottom of our 'news' page.
Discrepancies
(1) Between MN83803AL and -AK diagrams:
The pin designations on the camcorder repair manual supplement for the PV-950-B and PV-950B-K camcorders are the same as those in the MN83803AK data sheets supplied to me by Tom Cumming, and those scanned by Jep Holtrop and presented elsewhere on this web site, with the following exceptions:
(1) The /VCP and VCP data lines are transposed (switched).;
(2) The /HCPx and HCPx data lines are similarly transposed;
(3) The order of the HCP1 through HCP4 data line pairs (HCP1 and /HCP1 being an example of a pair) is also reversed; e.g. HCP1 and /HCP1 in the camcorder schematic are at pins 33 &32, the same pins that on the MN83803Ak data sheet are designated /HCP4 and HCP4, respectively.
These may be nomenclatural differences only, which, if so, bolsters the claim that the MN83803AL is pin-compatible with the MN83803AK.
Furthermore, you may notice that the pin mappings from the MN83803AK to its LCD match those from the MN83803AL to its LCD (see numbers in parentheses on schematic with MN83803AK, which I presume is used to indicate this mapping).
(2) Between voltage readings on the schematics:
You will readily notice that some of the voltages printed by the pins on the MN83803AK in the PV-D705 series repair manual, linked above on this page, do not agree (e.g. pin 17, 'Vdd', tied to the 4.5 volt supply rail, reads '2.3').
(3) Discrepancies not visible on the details of the schematics, as presented here:
Supplementary information: Both camcorder schematics that I've found showing the IR3y05y in use as the video processor consistently show a sync tip voltage of +0.6 volts, both at the source (the IR3y05y's sync separator output) and destination (SYNC/VIDEO pin 6 on the MN83803Ax). This is exactly consistent with the IR3y05's data sheets, which peg the sync separator output 'on-state' voltage at 0.6 volts max (no min or typ are given for this value).
The MN83803AK manual gives min, typ, and max voltages for VIDEO input only (all the camcorder schematics we use have the chip's pin 6 (SYNC/VIDEO) in 'SYNC' mode instead): These values are 0.7 (min), 1.0 (typ), and 1.7 (max) (all in volts, peak-to-peak). This differs from the application we see in our camcorder schematics (0.6 volts from system ground) in SYNC mode, so it cannot be expected to apply to pin 6 when it is in SYNC mode. (Note also that volts peak-to-peak does not necessarily say where the signal is with respect to system ground).
For more about sync voltages, check out the data Julie S. Porter has collected on how to get the MN83803Ax to phase-lock to a video sync input.
(2) IR3y05y Supply Voltage (see above: Data sheet and camcorder schematic disagree) Amendment This discrepancy is only present on the circuit board assembly FPC connector labeling, not in the test voltage readings.
(1) Logic inputs
The PV940 & PV950 camcorders set VSEL 'low', while all the others set it 'high.' Since this controls VSYNC output from the MN83803Ax, which does not go to either the LCD or the backlight or to the IR3y05, or to any other subsystem on the MN83803Ax itself (it goes off the viewfinder circuit board assembly to do something somewhere else in the camcorder), it seems very unlikely that this difference is significant at all.
Those who are interested in onscreen display (OSD) applications for the LCD may find it useful to manipulate VSEL (as it seems that these camcorders may use the VSEL output to time their OSD character generator ICs). No other use for this switch seems readily apparent to me.
The function of PMODE remains to be explored. The MN83803AK data sheets plainly show that it controls the on-chip VCO oscillating frequency, and affects the HDATA rise position on the even lines. Quite clearly, the latter may have implications for what we are doing with the MCL0712A03 LCD, since HDATA is one of the LCD control signals.
Julie S. Porter and I have included jumper headers for this and other control lines on the most recent version of the controller p.c. board that we have designed-- so if our LCD is designed for a different HDATA rise position than provided for in the camcorders, we will be able to establish this by testing.
(2) Input conditioning filter for pin 5 of the MN83803Ax ('VIN' or 'V-AMP in')
The only difference in the structure of these networks themselves is the use of a 47k resistor at the input of the network in the PV-D705, as opposed to a 10k resistor at its input in the PV-950-B.
The VIN subsystem on the MN83803AK-- and presumably the other chips in this line-- has something to do with the vertical sync system; its output is modulated by the 'S/W' subsystem, which takes internal input from two points in the sync processing subsystems, as well as VSEL.
VSEL is held 'high' in both these camcorder series, so its function cannot have anything to do with the differing topologies of their input circuitry.
In other words, the filter network has nothing special to do with the setting of VSYNC's onset position, with respect to the NTSC sync onset (VSEL's function is to toggle this between one horizontal period and two).
The PV-940 and PV-950 camcorders do not send any inputs to pin 5 (VIN) of the MN83803AK/K chip, grounding it instead-- therefore do not have this filtering network at all. In these camcorders, pin 7 (SYNCOUT) is left open.
If noise filtering is the purpose of these networks in the other camcorders, maybe the single 1000pF capacitor from the IR3y05y's sync separator output to ground serves this function in the PV-940/PV950 series.
(3) Series resistor in the MN83803Ax's 'centering' trimmer network
(4) Power supply questions
It is also unclear, as noted in the 'care and feeding of the cybermaxx' power supply section , whether (1) the MN83803AK is really limited to 18 +/- 2 volts VDD2, and (2) whether the MN83803AL will run on 18 volts.
This is because internally consistent schematics for the camcorders that use these chips show the MN83803AK running on 14.1 volts, and the MN83803AL on 14 volts (the MN83803AK/K is seen elsewhere running, with an MCL0712A01 LCD, on 18 volts). In the case of the MN83803AL, there is a dissenting test point voltage of 18.1, but no support in the schematic for this number to show how it would get there.
Let us make a wild guess: that the MN83803A(K) data sheets we have been provided with are inaccurate, reporting the operating conditions (e.g. supply voltages) of the MN83803A, when in fact the subsequent chips in the line (MN83803AK, MN83803AK/K, MN83803AL) are able to work on a wider range of voltages than the plain 'A' variety.
The 'K' in the data sheets' 'MN83803AK' designation is pencilled in by hand. The most noticeable places where there are pencilled-in data, besides the title of the document, are on the chip package diagram, and on the Vertical System Output Timing and Horizontal System Output Timing tables. The column pencilled in is the column for our LCD: the 505 x 230 variety. (This makes me wonder if my MN83803A chip that Panasonic sent me will work in this format)
What if Panasonic was concerned mainly with updating the logic control tables for the horizontal and vertical subsystems when it updated the MN83803A data sheets-- and neglected to update the operating conditions and electrical characteristics sections?
This would explain how we can see an MN83803AK running on 14 volts, for example.
It would leave the question of the operating voltage of the MN83803AL in question. But we know that in the Cybermaxx, an MN83803AK chip ran with this LCD (thus needed 18 volts VDD2); that the MN83803AK/K runs on the same voltage; and that the PV-D705 camcorder is supposed to have an MN83803AK running at 14.1 volts. Therefore let us guess that the MN83803AK will run on a wide spread of voltages, encompassing 14 and 18 volts.
It is not unreasonable to guess that the MN83803AL is similarly versatile; but neither is it a given.
(5) Interpreting the discrepancies noted earlier on this page
(6) Family resemblances: Matching driver chips to LCDs
Two others use the MCL0512B01 and MCL0512B01M.
Those that use the MCL0712A01 run it on 18.1 volts, like ours, and run its LCCOM at about 5.08 volts-- very close to the same as ours (5.3 +/- 0.2 volts).
Those that use the MCL0512B01 and MCL0512B01M run them on 14 volts. LCCOM for the MCL0512B01 is the same as for the MCL0512B01M (see the voltage dividers feeding LCCOM: They give the same output voltages. 27k:4700+22K = 2700:470+2200). Their LCCOM is much closer to the video center voltage than the MCL0712A01's and MCL0712A03's is.
Thus we can see a family resemblance within the MCL0712s, as within the MCL0512s, according to their respective supply and LCCOM voltages; and we can readily distinguish the two families from one another.
Without a more comprehensive set of data sheets for the MN83803Ax series of chips, and for the other LCDs, it's a matter of educated guesses as to what conclusions we can draw just by looking at schematics. It seems plain to see that there is something different about the VCO on the MN83803AL, since its 'centering' trimmer network differs from that of the AK and AK/K; it seems obvious that the control signals for the MN83803AL are the same as those of the AK and AK/K (the same settings are used for LCDs of 505 x 230 configuration); it seems clear that the AN2523FAK/K provides a different sync signal than the IR3y05y (different test point voltages, different signal to MN83803Ax's pin 5 (VIN)); Panasonic Service Company lists the MN83803AL as a direct replacement for the MN83803AK; Panasonic packs a chip labeled 'MN83803A' in a box labeled 'MN83803AK'....
So we see a family resemblance between the chips, and a consistency in their treatment as interchangeable parts by Panasonic. However, we see the MN83803AL listed in a parts list for the Cybermaxx 180k (which used a 180kilopixel LCD, not a 116,150-pixel device like ours). Does this chip have extended functionality not listed for the MN83803AK in its data sheets? If so, how is it implemented by the chip's control logic? (presumably, if it has extended functionality, its implementation does not conflict with that of the functions we would use). Does the -AL sacrifice its PAL modes or another NTSC mode of operation to make use of the input designation those would use? Or does it use a pin that is unused on the -AK (e.g. pin 23) to effect a change of mode?
Similarly, as noted above, inconsistencies between the MN83803AK data sheets and the schematics, and inconsistencies in the schematics, leave questions about the VDD2 the various MN83803Ax chips use.
Inference can only get us so far. The camcorder schematics show a few configurations that presumably work, but we cannot tell without empirical study just how the various devices may be interchangeable or how their variants (IR3y05/y, MN83803A/K//K/L) differ in function and/or electrical characteristics. This suggests that to get a working circuit at first, we should stick as closely as possible to the schematic of the camcorder.
I will be mounting an IR3y05y, supplied by Panasonic, and an MN83803AL chip in my implementation of this PV-950-B design, for this reason.
Power supply questions about the IR3y05y are easy to resolve, since if it won't run on 12 volts, all it will take to make it work will be to switch in a different kind of voltage regulator. While there is no provision on the board for this, this is not too complex a physical design problem to solve.
Our present circuit board design, revision 'L', incorporates jumpers for the logic input pins of the MN83803Ax, and a configuration of vias (holes connecting one side of the circuit board with the other) that allows it to be configured with an input network like the PV-D705 camcorder (as in Kristian's design), or like the PV-950-B camcorder. This will give us some freedom to test the various designs.
Related manuals
From my notes:
PV-D705-D
Note that since this was from a telephone conversation, the dashes shown above may not be in the right places and probably should not be used in reference to these camcorders when asking for manuals, et cetera, in case this confuses a database query.
RGB Signal Conditioner
RGB Signal Conditioner Schematic Right-click to download.
The inverter uses the Comlinear (National) CLC 430's blanking feature to switch between an inverting and a noninverting video amplifier.
A zener diode sets a -14 volt reference, divided down to -7 volts by a precision voltage divider.
An inverting summing amplifier at the output stage mixes this -7 volt signal with the inverter's output, resulting in an output signal that alternates around +7 volts depending upon the state of the FRP input.
The test circuit uses an LM301 which was handy, far from an optimal choice. Any amplifier that runs on +/- 18 volts (or >= about +/- 15 volts), and will work stably at a gain of -1 at video frequencies, will be preferable.
Lacking a stable power supply for now, I haven't tested this design to run the LCD.
If a noninverting summing amplifier is used for the output, the associated +7 volt zener reference could be used as a reference by another voltage divider to yoke the LCD's LCCOM voltage to the video center voltage directly.
Alternatively, another zener reference voltage could be used to set LCCOM at a stable voltage.
The present design on Kristian Bognaes's driver board uses a resistor voltage divider to set LCCOM, relative to the power supply voltage. This is the same setup as is used in the camcorders. With a stable power supply voltage, this should work fine.
Click here to go back to the main LCD project page
Click here to go back to my home page
email: pstewart+atsign+gwi+dot+net
(1) One I've caught is on the input to pin 6 of the MN83803AK (SYNC/VIDEO), which reads '0.2', whereas this line's origin, at the AN2523FAP/K video signal processor's sync separator output (pin 23 on that chip) reads '1.3'.
All of the camcorder schematics I've obtained show the the LCD driver with H1, H2, V1, and V2 control lines held 'Low' (at ground); and VMODE held 'high' (at the positive digital supply rail)-- consistent with their use as drivers for 505(H) x 230(V) LCDs. The MN83803AK data sheets specify exactly this control configuration for 505 x 230 (see their Horizontal System and Vertical System Output Timing tables for these values).
A close inspection of the schematics for the PV-D705 series camcorders, and those of the PV-950-B series camcorders reveals that both the MN83803AK and MN83803AL use very similar input conditioning networks for their pin 5 ('VIN'); however, the PV-950-B uses this network to connect Pin 7 (SYNCOUT) to pin 5 (VIN or 'V-AMP in'), whereas the PV-D705 uses it to connect the video processor's sync separator output to pin 5 (VIN...), so that pin gets a filtered version of what pin 6 (SYNC/VIDEO in) is getting. The difference that might motivate this is the use, in the PV-D705, of the AN2523FAP/K video processor to provide sync input, and use of a different chip, the IR3y05y, for this purpose in the PV-950-B. I.e. they are set up differently because something is different about the sync signal they are conditioning (one gives pin 5 a version of the sync the AN2523FAP/K is sending to pin 6, while the other gives pin 5 a version of what the MN83803AL has done with the sync pulses the IR3y05y is sending to pin 6; the MN83803AL has already processsed the latter before sending it, through this filter, to pin 5).
The series resistor used with the 'centering' trimmer in the MN83803Ax's VCO system differs between designs that use the MN83803AL and those that use use the AK or AK/K. Those that use the AL (PV-950-B series) use a 10k series resistor; those that use the AK or AK/K use a 4700 ohm resistor here.
Some questions remain about the differences there may be between the IR3y05 and IR3y05y chips, and between the various flavors of MN83803A. Since the schematics are fairly consistent regarding the sync 'on-state' voltage the IR3y05 chips supply, the main functional questions we have now concern the proper supply voltage for the IR3y05y (viz, whether it can run on a convenient 12-volt supply or will need something else).
In light of the consistencies and parallels noted, it seems that all the discrepancies we have pointed out in the schematics (above, on this page) are either trivial typographical errors or variations in labeling practice, with no functional implication we need to worry about.
Two camcorders I've gotten schematics for (the PV940 & PV950) use the MCL0712A01 LCD (remember ours is the MCL0712A03).
For those interested in obtaining camcorder manuals, the following is a truncation of the list I was provided with by the Panasonic literature search people (I didn't write the whole, orally quoted, list down), of camcorders which, according to my note anyway, apparently all use the MCL0512B01 viewfinder LCD. Anyone desiring to supplement this information can of course call the Panasonic numbers listed elsewhere in these web pages. When I called, Panasonic's telephone people were both courteous and helpful, though their factories refused to provide the I.C. specifications I requested, to safeguard against reverse engineering of the chips.
PV-ICQ-575
485 (?)
PV-IQ-375
PV-D075
PV-ICQ-375D
PV-ICQ-525
PV-ICQ-325
PV-ICQ-315
(PV?) VM-555
Here is a design for an RGB inverter, switchable by the frame rate pulse (FRP), followed by a level shifter that adds this signal to +7 volts, which is the LCD's rated video center voltage. It is provided to supplement Kristian Bognaes's driver board, which lacks a video processor.
