Repair "Pentagon 48"
ZXNet echo conference «hardware.zx»
Hello jtn
such a question: on the emulator in DIZZY UNDEGRAUND, the stripes on the border coincide with
main screen.
In real life they are located slightly lower.
http://almaz.homesait.net/upload/DSCF0009.JPG (360kb)
The demo says that the duration of INT should be Pentagon. But I have
Pentagon (48 modified to 128)...
How can this be cured?
As far as I understand, the duration is set by the RC chain C3 (1H) and R12 (100k).
How can you quickly select the desired denominations? (no oscilloscope :( )
Hello Almaz
Alm> As far as I understand, the duration is set by the RC chain C3 (1H) and R12
Alm> (100k).
Alm> How can you quickly select the required denominations? (no oscilloscope :( )
Depends on when the impulse occurs - during the active screen or
reverse motion. According to the offhand scheme, INT passes immediately after BLANK.
This means that if it were submitted instead of one of the RGBs, the length of the strip would be
visible. Pixel - 143 nS. The only problem is that the signal does not have a buffer, so
that connecting directly is not suitable. But if you take a CMOS buffer, it will pass.
74HC04 or something similar. But not TTL!
Hello Almaz
Alm> How can you quickly select the required denominations? (no oscilloscope :( )
In the ZX-Review in one of the last ones for the 91st year in the “Forum” section this was described
focus. There is this file on trd.speccy.cz. If you can’t find it, drop me a line, I’ll
I’ll scan it for you from mine.
Hello icebear
icebear
found in the file ZX-Review-1991-01-12.pdf on page 556:
> 1) Gradually reducing the capacitance of the capacitor, determine the threshold for
> which
> keyboard polling stops. When the computer is turned on, it displays on the screen
> "(Since) 1982...", but
> does not respond to key presses.
> 2) Increase the resulting threshold value by about 1.5 ... 2 times
> and solder into
> circuit capacitor of the required capacity.
Is this it, or are there other methods?
Hello Almaz
int has a pulse length - it affects the performance of some programs (ei:ret
which do) optimally 28 cycles, measured by the same Kovalevsky test
And there is a phase - a displacement relative to the CS or the beginning of the pixel screen (at
Pentagon 80 border lines on top), that’s what you need for multicolor effects
select this phase, unless of course there are 224 bars in line as on penta128 and there are no
whites =)
Hello, volton
vol> I dug up Pentagon48, but with one bug, namely: The cursor does not blink
In the Flash driver is D16 (K561LH2 pins: 1, 2) VT2 (usually
KT315). Sometimes for stable operation it was necessary to select R16 less often than R19. Overall
The circuit of the C driver is unsuccessful.
Alternatively, an additional double counter can be installed
(divisor by 32 or 64), send it to the counting input INT, and get from the output
high-quality Flash.
Hello jtn
I dug up Pentagon48 expanded to 128, almost
working, but with one bug, namely: The cursor does not blink,
at the same time, the keys work, int-norm...
Plz, tell me where to start dancing.
It seems like a small thing, but not pleasant :)
Hello Conan
Finally, I figured out what kind of Pentagon I have. This is 48
modified to 128. The board is covered in wire noodles, it’s scary to look at it,
touching with your hands is even worse. However, there is a need to touch it, so please
respected colleagues, please refer me to the primary sources: diagrams of the Pentagon-48 and its
modifications up to 128.
Specifically, on my board they are soldered with a sandwich: on D1 (KP11) - another KP11, on
D5 (LL1) - LL1, on D41 (TM9) - TM9 (see the wiring diagram posted above), and well
additional 8 pieces of RU5 are added to the existing ones.
Among the features is the Profi firmware (after resetting, “ZX Spectrum.RAM JV” appears
Kramis v02 - can anyone share a description of the add-on? menu items - I will
grateful) in 27512 and a proprietary interruption (what is the correct name for this, not
I know) - in demos where there is multicolor adjustment, I have to choose
128/Scorpion, then everything is fine. As a result, Shock doesn't work normally for me.
Megademo is neither original nor fixed for the Pentagon.
Hello Lennon
len> People, what are we talking about? :v2_blink;
len>; As far as I remember, they sealed the round window of sectors at 5.25 so that
len> floppy disks formatted in Speck could be read on a PC.
len>
len> What kind of know-how is this? Why am I interested, I have a project TR-DOS for
len> Spectrum +3, I want to use a 3.5 drive there...
len> I thought that there was no need to glue anything :v2_blink; .;
Floppy disks for 720 and 1.4 are distinguished by the presence of a window in the latter. HERE it is
They seal it so that the drive thinks that it is a 720 floppy disk.
Hello, CHRV
CHR> Floppy disks for 720 and 1.4 are distinguished by the presence of a window in the latter.
Where is it located?
And that 1.44 won’t work on Tr-Dos? :v2_wacko:
Hello SMT
SMT> and if you give +5 instead of 12, it won’t work (although it still won’t work
SMT> will be)
SMT>
SMT> "sealed" HD 3.5 floppy disks, by the way, really worked for me
People, what are we talking about? :v2_blink;
Ha; as far as I remember, they covered the round window with sectors at 5.25 so that you could
It was possible to read floppy disks formatted on Spec on a PC.
What kind of know-how is this? Why am I interested, I have a project TR-DOS for Spectrum
+3, I want to use a 3.5 drive there...
I thought that there was no need to glue anything :v2_blink; .;
Hello Lennon
len> What is the problem?
To read 1.44 VGB you need 2 MHz. Well, plus software support...
Hello Lennon
len> Where is it located?
len> And why won't 1.44 run on Tr-Dos?
yayyy, as always, you are on time and on topic.
on 1.44 floppy disks (HD) there are two holes: one with a latch - write protection, the other
(opposite) - a sign that the floppy disk is HD. It is not available on 720kb floppy disks (DD).
Accordingly, on the Spectrum in 99% of cases they use DD and so that the drive
worked correctly, you need to seal this hole (as an option, solder the microphone into
disk drive). but some craftsmen managed to remake the controller and force
VGshku read HD disks :)
Hello, The Exploited
The> yayyy, you are on time and on topic as always.
To the topic, this was discussed here, I answered with a question. :v2_tong;
The>; on 1.44 floppy disks (HD) two holes: one with a latch - protection
The> records, another (opposite) is a sign that the floppy disk is HD. at 720kb
The> floppy disks (DD) do not have it.
Thanks, but I already guessed it myself. I just, by analogy with 5.25, was looking for a window for
magnetic disk, not end-to-end...:v2_yahoo;
The>; but some craftsmen managed to remake the controller and make
The> VGshku read HD disks
What is the problem? :v2_blink: