Scorpion ZS 256 Turbo+ (diagram)
ZXNet echo conference «hardware.zx»
Hello Orionsoft
Is there no installation?
Hello deathsoft
dea> remove the zip extension because the site does not allow uploading tif
and pack tiff in zip?
=)
Hello Orionsoft
Here is the installation (remove the zip extension, because the site does not allow uploading tif,
had to add zip)
File: sc256_turbo+.pcb.tif.zip http://zx.pk.ru/attachment.php?attachmentid=3259
Hello Orionsoft
At the request of the workers!
Gentlemen, comrades, Scorpios, it’s weak to create a website (schemes, special programs) according to
Scorpa?
PS; TO; question about Turbo-FPGA. While studying the topic, we managed to find out that the firmware and circuits
there were at least two inclusions - for modification of the yellow Scorp and standard for
green.
File: ScorpionZS256.zip http://zx.pk.ru/attachment.php?attachmentid=3260
File: ScorpionZS256Turbo+.zip http://zx.pk.ru/attachment.php?attachmentid=3261
Hello spensor
spe> PS; TO; question about Turbo-FPGA. While studying the topic, we managed to find out that
spe> there were at least two firmware and switching circuits - for modification of the yellow one
spe> Scorpa and standard for green.
My turbocharger circuit was the same as in turbo+ (green). The chip was Intel
85c220, produced in 1986 (later altera epc220 started making them).
Turbocharging was done by the company Scorpio (they had such a service). Turbocharged
consisted of soldering this epld onto the breadboard field and soldering it with MGTF wires
to different places in the diagram.
The yellow scorpion is apparently some kind of antiquity (apparently sc11 boards or even more
ancient) (my scorpion has a gray board, without a green mask, board number
sc14). The first scorpions were made according to some simplified scheme, where many
the delays were set using RC chains; there is nothing like that in the turbo+ circuit.
The only difference between my scorp and turbo+ is the absence of a 5V->12V converter for
power supply VG93.
Hello deathsoft
dea> Turbocharging was done by the company Scorpio (they had such a service).
dea> Turbification consisted of soldering a given epld and to the prototype field
dea> soldering it with MGTF wires to different places in the circuit.
I apologize, but this clarification does not clarify anything... I said that
there were at least two “fillings” of the FPGA, and, accordingly, two circuits for connecting the FPGA!
The difference in the switching circuit of the microcircuits is insignificant (2 signals), but this is no longer
allows you to simply replace the FPGA with Turbo firmware between green (more
new) board and modified yellow (old) one.
Hello spensor
if you insert a PLM from a yellow scorpion into a green one, then everything works =)
only we don’t see pixels on the screen (there is a border) and the screen is white
Hello Orionsoft
Differences between alter signals for the yellow board:
3 (/RAS)
4 (/TRB_OFF) turn off turbo
5 (/TRB_ON) turn on turbo
15 free output
Remains unclear (contacts):
4, 12, 14, 17, 19
The rest is like green
Hello deathsoft
there should be an input for the turbo button
Hello deathsoft
Both firmware sources were obtained by decompiling the jed file. Firmwares are calculated
to the "green" board version SC15. Firmware comparison for equivalence is not
was carried out (the equations may generally turn out to be equivalent).
The original .jed files are also posted here
File: sc15jed.rar http://zx.pk.ru/attachment.php?attachmentid=4669
Hello, Evgeny Muchkin
Evg> and on SC16 there is
No, there is only SC15.4 for epm7032 (but it cannot be decompiled)
Hello deathsoft
Thank you.
Optimization or face: v2_cheer:
Hello Orionsoft
deathsoft, is it available on SC16?
Hello deathsoft
what kind of utility is jed2ahdl?
what chips (jed) are supported?
Hello Orionsoft
Ori> if you insert a plm from a yellow scorpion into a green one, then everything works =)
Ori> only we don’t see pixels on the screen (there is a border) and the screen is white
Actually, it cannot work correctly, because... there are differences (see post #38)
Hello deathsoft
dea> jed2ahdl - Atmel utility. Only PLM type 16v8 22v10 are supported
dea> 85c220, etc. CPLD (Tim Max and Xylinks) are not supported.
I went to xsay and something is not downloading
maybe you can throw it at me
Hello Orionsoft
Ori> what kind of utility is jed2ahdl ?
Ori> what chips ( jed ) are supported?
jed2ahdl is an Atmel utility. Only PLM types 16v8 22v10 85c220 and
etc. CPLD (Tim Max and Xylinx) are not supported.
Hello, CHRV
Apparently they were offline
and now, according to the subject
who makes the TM9 signal that is on the 9th leg of DD57?
Hello Orionsoft
Ori> I went to x say but something is not downloading
Ori> maybe you can throw it at me
http://www.atmel.com/dyn/resources/prod_documents/JED2AHDL.ZIP
Hello deathsoft
deathsoft, orionsoft, colleagues, can you find VG turbocharging schemes for
yellow and green scorpions? Best of all, tell me how to convert BDI to
yellow with its turbo so that it becomes the same as on green. :)
I'll bet the beer! A lot! :)
Hello Orionsoft
Ori> apparently they were offline
Ori>
Ori> and now, according to the subject
Ori> who makes the TM9 signal, what's on the 9th leg of DD57?
It is made by the DD19 decoder (leg 12). (TR-DOS port decoder)
Hello deathsoft
dea> It is made by the DD19 decoder (leg 12). (TR-DOS port decoder)
KP16???
Hello Orionsoft
Ori> KP16 ???
ID7 (DD19), where there are 2 diodes (legs 10 and 14, respectively). In the yellow board
This decoder is also available (DD54), the connection circuit is similar.
Hello, Evgeny Muchkin
Evg> Can you elaborate on this point?
The formation circuit is made on 2 triggers (DD59-1 and DD9-1) and an AND-HE element
(DD69-1), as far as I understand, the AND-HE element was taken from free ones on the board (for
yellow DD57.1 or DD57.2), TM2 is installed on the prototype field next to VG93.
Next, a piece of VG93 comes off and looks at this circuit, and they also turn on it
VG93 signals that have not been used before, the output of the circuit is turned on
VG93 (see green diagram).
Hello, Evgeny Muchkin
Evg> Are the bdi circuits themselves the same in yellow and green?
Almost identical, only the 12V power supply circuits of the VG93 differ, and the circuit
formation of CLKWG.
Hello deathsoft
dea> and CLKWG formation scheme.
Can you elaborate on this point?
Hello deathsoft
dea> Turbification is done on the TM2 chip, which generates the signal
dea> CLKWG.
And all this is the same for both yellow and green? And the bdi circuits themselves
Are yellow and green the same?
Hello, Evgeny Muchkin
Evg> deathsoft, orionsoft, colleagues, can you find out the VG turbocharger diagrams
Evg> for yellow and green scorpions? And best of all, tell me how
Evg> remake the BDI on yellow with its turbo so that it becomes the same as on
Evg> green. :)
Evg>
Evg> I'll bet the beer! A lot! :)
So the turbochargement of the VG93 does not depend on the turbochargement of the scorpion itself. Turbocharged
VG93 consists of supplying double frequency pulses to VG93 during
execution of the seek command (moving to the desired track). Turbocharging is done
on the TM2 chip, which generates the CLKWG signal. In yellow scorpio on VG
a 1 MHz signal is supplied directly from the divider counter (DD56), and in green
a 2 MHz signal is used from the same counter (DD65 according to the green circuit), in
depending on the operation performed by VG93, this signal is either divided in half, or
served without dividing. The divider is made on TM2 (DD55.1 and DD9.1 according to the diagram
green).
Here is another description of the modification of yellow to green (it also talks about turbocharging the VG93
written). Description provided by Scorpion. Mikka did this modification
in his yellow scorpio, incl. he can tell it better.
File: TURBOSC.rar http://zx.pk.ru/attachment.php?attachmentid=4696
Hello Orionsoft
Ori> well, I understand, this one is the DD65 =)
Yes, it turns out I have half of the diagram printed (old) from Orcade
positional designations are mixed up. I checked the diagram in Orcade, everything is correct there.
I'll have to throw out that old printout so there's no more confusion.
Hello, The Exploited
The> do you actually have any processors that start in it?
Yes. All other processors that I have work fine in it. In that
including Goldstar at 4 megahertz, Zilog at 4 (they don’t really like current turbo), and
kmos Z84C006/10 pec work great in this yellow board.
The problem lies with these three.
Hello, The Exploited
The> an oscilloscope is obviously needed
Where should they look?
Hello deathsoft
deathsoft, orionsoft, I come to you again with a petition about green and yellow ;)
I have 3 processors, completely new: Z0840006PSC (2001).
None of them want to start, being stuck into the yellow turbo board. :(
I take out the processor, ROM; I plug them both into the green board and voila! the computer started, that's it
It works great both with and without the turbo, whatever you like, in general.
You guys are bright, maybe you can tell me what can be done with the yellow board so that
to liken the latter to green and still be able to launch these capricious
Z0840006PSC ?
Hello, Evgeny Muchkin
Evg> Not one of them wants to start, being stuck in a yellow turbo
Evg> board.
But in general, do you have processors that start in it?
Hello, Evgeny Muchkin
You obviously need an oscilloscope
Hello deathsoft
check the resistor value between the 6th cpu leg and +5V!
680 OM!
Picture: http://img502.imageshack.us/img502/9462/z80bc2.gif
Hello, Evgeny Muchkin
Evg> I have 3 processors, completely new: Z0840006PSC (2001 onwards).
That in 2001, Zilog was still releasing NMOS processors?
Most likely the CLK signal level is outside the passport mode for NMOS
protsov. For NMOS, the CLK signal should have a swing of almost 5V, level 1tsy
must be at least 4.8V
Hello deathsoft
dea> Most likely the CLK signal level is outside the passport mode for
dea> NMOS processors.
exactly like that
Hello, CHRV
CHR> Atm-turbo costs 470 ohms
And it’s the same at General Sound.
Hello, Evgeny Muchkin
deathsoft, Orionsoft, The Exploited, thanks for the thought!
Although I haven’t checked it yet, it seems that the problem is indeed in raising the 6th leg to +5V.
I looked at the yellow and green diagrams:
- on the yellow one I didn’t find a pull-up resistor at all: v2_blink;
-; on the green diagram, it is designated as R33 with a nominal value of 1K5.
When I looked at the green board, I discovered that the rating of R33 is 560 Ohms (not
1K5) in total, so apparently the processors work on this board.
I tested the same percentages on my freshly harvested green Mikka_A; he doesn't have them
earned. And on his board R33 is “correctly” as per the diagram and
installation - 1K5. But to be happy, apparently, you still need a lower denomination. :)
I'll check it and write back! Thanks again!
Hello, Mikka_A
Mik> if I understand correctly, kmos is also distinguished by competent processing
Mik> undocumented commands, unlike nmos, right?
Quite the contrary. By installing nmos for yourself, you get more compatibility with
previously written software.
In kmos, some undocumented commands do not work the same as in
nmosakh, and, as a result, we get less compatibility with older software.
Mik> that is, the ideal option is
Mik>
Mik> Z0840006PSC
Mik>
Mik> if it works stably at Turbo frequency.
Almost like that :) The main thing is that the percentage is not Z84C.....although with such a processor
you may not feel any discomfort at work. :)
Hello, Mikka_A
Mik> that is, numbers 84 and 084 are monopenisular? is there a zero there.. no...
Exactly :) The main thing is not Z84C... or Z084C... etc.
Hello deathsoft
dea> That in 2001, Zilog still released NMOS processors?
Yes, here it is in the photo, the one at the bottom:
http://zx.pk.ru/attachment.php?attachmentid=4182
Hello, CHRV
CHR> Atm-turbo costs 470 ohms
In KAE, Nemo generally has a choke and a resistor with a small value.
Hello, Evgeny Muchkin
Evg> Almost like that :) The main thing is that the percentage is not Z84C..... although with this
Evg> you may not feel any discomfort while working with the processor. :)
that is, numbers 84 and 084 are monopenisular? is there a zero there... no...
Hello, Evgeny Muchkin
Evg> I tested the same processes on my freshly harvested green Mikka_A; they
Evg> did not work for him. And on his board R33 is “correctly”
Evg> as per the diagram and installation - 1K5. But for happiness, apparently, you still need
Evg> lower denomination. :)
Evg>
Evg> I’ll check it and definitely write back! Thanks again!
And I'll check. Just first, just for fun, I’ll poke it with an oscillator and see the scope.
And I'll see what's on my turbocharged yellow one.
Because the prots worked there.
Hello, The Exploited
The> exactly like this
If I understand correctly, kmos is also distinguished by competent processing
undocumented commands, unlike nmos, right?
that is, the ideal option is
Z0840006PSC
if it works stably at Turbo frequency.
Hello, Evgeny Muchkin
Evg> When I looked at the green board, I discovered that the rating of R33 is 560
Evg> Ohm (and not 1K5) in total, so, apparently, there are processors on this board
Evg> work.
Atm-turbo costs 470ohm