Pentagon renovation — ZXNet «hardware.zx»

From Dobrynin Sergey To All 12 December 2005

Hello,geners How does it respond to reset?

From Nikita Alexandrovich To All 12 December 2005

Hello, All I am repairing a Spec Pentagon 128 (circuit like 1991) according to the official version, RF2 burned down (that itself is a bit strange IMHO) on the spec, diodes were bitten off (well, barbarians) to tighten the colors to intensity (for a long time I couldn’t understand why there were 3 connectors on the board not connected to anything other than X1) The most interesting thing is that the wires came to the system connector and on the back wall I soldered the cable for the scart - surprisingly the image is quite clear Naturally, before turning it on, I pulled out all the devices from the cribs (CPU, ROM, FDD) When you turn it on, a “board” appears on the screen, as it should be but part of the screen (top 10-15cm) is overexposed compared to the rest the board itself has a twist - those top and bottom cells are 2 times narrower than the rest on the board itself there are artifacts in the form of constant and blinking squares in 1 familiar place When installing the CPU, vertical stripes will appear (again as expected) it looks like || || || || etc. when installing 27512 the screen is completely blank The RUs don’t seem to heat up, which does not mean they are working properly On the board itself, some diodes have the same resistance in both directions :( (I don’t think they’re broken, but I’m thinking of replacing them) The K555Ir23 (DD55) gets very hot - after 2 minutes you can’t lift your finger Threat, please don’t give answers like “look for all this in the forum already” discussed" how some people act :mad:

From Nikita Alexandrovich To All 12 December 2005

Hello, do_se no way, but this is not a fact at all, I haven’t fully figured out what else is there, but what's missing With the process installed, the strips move left to right one familiar place

From Vasily Strochkin To All 12 December 2005

Hello, do_se Perhaps the twists already occur on the TV? Just very lucky for them situation (bright raster). If the oscilloscope shows a constant change in signals on the address/data bus, then There should be no overheating of the microcircuits. Most likely in this chip and problem. You lubricate the memory with alcohol and it resets; if it starts, it’s the chip that was wetted was beaten. The method is very, very unreliable, but for RU5 it worked very well not bad. The rest of the chips can do the same, but for now some of them are overheating There is no point in doing such a search.

From Dobrynin Sergey To All 12 December 2005

Hello,geners It looks a lot like the computer is starting, but either you don’t see it because the screen is “broken”, or it “freezes” at the very first instruction.... Look for an oscilloscope and look at the signals on the address bus and data bus, on the processor, ROM, RAM, maybe misses something... But most likely everything is in order there (States change, without processor, without ROM, with ROM)... By the way, which RF2 were you talking about? I think there's only there was one ROM... Try to figure out the screen. It is quite easy to calculate on the counters frequency (according to the diagram). Since your screen is turned up, the frequencies are definitely "leaving". Try to look for breaks, sticking to ground, +5V (looked for by the tester, although more convenient with an oscilloscope), short circuits between tracks (clearly visible on the oscilloscope the third state, as on the data bus, is normal there), I think You'll find the problem pretty quickly. After sorting out the logic, then everything is usually he takes off at a full swing, unless of course there are bad RAMs, but he will definitely strategize should. The usual scenario is reset - black screen - white screen + white border - menu. By the way, it can start in 48 mode, then if the bottom of the screen is “wrapped” You won’t see anything, as you remember, there’s only the inscription “Sinclair.....

From Nikita Alexandrovich To All 12 December 2005

Hello,geners I think volvulus is excluded on TV, I connected the delta with the same symptoms so her screen is brighter and there is no torsion

From Nikita Alexandrovich To All 12 December 2005

Hello jtn so there is no programmer RF2 is how I understood the lprint subprocedures, it stores a capacity of about 8 kb The screen is only wrapped (visible) without any process with the percent there is a reaction to the reset - this is exactly the walking left and right you click once - click left 2 - click right (not by itself) Black screen when installing ROM The image is generally very dim, I had to curtain the windows

From jtn To All 12 December 2005

Hello, vasstr D55 - what's on the printer port? look for short circuits at the outputs (you never know...) and bite it boldly, she’s probably throwing SD. I recommend flashing the test and pasting it instead of 27512 - the test is on the forum

From jtn To All 12 December 2005

Hello, vasstr D55 - what's on the printer port? look for short circuits at the outputs (you never know...) and bite it boldly, she’s probably throwing SD. I recommend flashing the test and pasting it instead of 27512 - the test is on the forum

From jtn To All 12 December 2005

Hello,geners kd522 usually. Over the years, diodes had to be replaced only when they punched 12 or 5V - I don’t remember a single case where they just flew out - forget it. It’s better to look for a stitched collar, and in Yaroslavl and especially in Rybinsk there are still real people and I think hardware people who can help, you just need a little search..

From jtn To All 12 December 2005

Hello,geners gen> without process gen> with the percent there is a reaction to the reset - this is exactly the walking from left to right gen> click once - left click 2 - right (not alone) This is all normal - this is how it should be if the processor and memory are working properly. Poochka died with a high degree of probability. Regarding the inversion - most likely it’s a matter of signal amplitude, as already written here.

From Nikita Alexandrovich To All 12 December 2005

Hello jtn I agree, and I know the setup procedure after the side (specifically the Pentagon) :) I'm worried that the memory chips burned out for a reason - for example, 2716 in the city it costs 52 re - IMHO a little expensive, not to mention 27512 in Prince 16 I have 4 of them - you can experiment, I’ll replace DD5 For sure, I’ll change all the diodes so that you don’t think about them Maybe someone can download the flashed ROM, I’ll send it in mikrukhs (conditionally empty)

From Nikita Alexandrovich To All 12 December 2005

Hello,geners As far as I understand, it doesn’t matter which diodes to use or d220 or kd143, it’s probably better not to :) ;)

From Dobrynin Sergey To All 13 December 2005

Hello,geners gen> and so you tell me if the ROM failed, why couldn’t the diodes gen> the mikruha actually BURNED, even the window burst

From Dobrynin Sergey To All 13 December 2005

Hello,geners I looked at the diagram of Pentagon 128 (look there http://zx-f.ru/sch/pentagon_128.zip) diodes should not float at this stage, because they cost: 1. on the tape input (by the way, they are turned towards each other, by definition will call in both directions) 2. on address lines of the keyboard 3. on mixing color signals 4. it is also in the drive controller, but this is not important, you can even use VG tear it out at the first stage. That is, diodes cannot light up a computer in any way.

From Nikita Alexandrovich To All 13 December 2005

Hello, do_se do_> 3. on mixing color signals maliciously bitten :) do_> 4. it is also in the drive controller, but this is not important, you can do_> generally rip out the VG at the first stage. I don’t notice it at all at first do_> 1. on the tape input (by the way, they are turned towards each other, according to do_> by definition they will call in both directions didn't even pay attention to them besides them, there are still some soldered somewhere apparently someone was finishing something, some of the little things are not the same as in the diagram, that’s why First I’ll figure out what’s fancy and then I’ll solder

From Nikita Alexandrovich To All 13 December 2005

Hello, do_se yeah, let’s turn it over, it worked for five years without problems and used it actively - then died

From Nikita Alexandrovich To All 13 December 2005

Hello jtn Yes, I know real people, sometimes I sometimes, but as for Rybinsk, it’s generally a pipe :( and tell me, if the memory flew out, then why couldn’t the diodes The little thing just BURNED, even the window burst :)

From Nikita Alexandrovich To All 14 December 2005

Hello,geners By the way, while people were connecting the spec to the ega monitor, they thought, why not? do not make an EGA_CARD from the board, you only need to take 8 wires and without modification the spec himself - give it up and use it

From Andreas Kaiser To All 16 December 2005

Hello,geners gen> 3 don't stuff me with diagrams, I have them on paper of much better quality We went among the people since the quality is good. gen> 4 caro is probably right that the RF is for the printer port Kamil is right without “probably” :)

From Kamil Karimov To All 16 December 2005

Hello, vasstr vas> I have absolutely no idea about the contents of MS RF2, but I have suspicions vas> that without the code located in this chip you will not run anything. This ROM contains the ZX-Lprint II driver for the printer to operate. You can safely pull it out, especially since the printer Apparently, it still won't connect.

From Vasily Strochkin To All 16 December 2005

Hello,geners gen> not a resistor but a mikruha ir23 m-ya with domestic mikruhas is generally weak gen> sign gen> but it doesn't look like a resistor matrix gen> and it’s impossible to connect a printer while it’s running, because the computer is locked gen> in the case is something between a desktop top and a tower and to get to gen> remove the port on the cover 555ИР23 = (SN)74LS374(N) KR1533IR23 = (SN)74ALS374(AN) They call this m/s register because its internal structure can be imagined (replace this m/s with) like 8 D-triggers. By voltage drop from logical 0 to logical 1 on leg 11, data is written to the register (inputs: 3,4,7,8,13,14,17,18). Data is output to output (2,5,6,9,12,15,16,19) when presence of logical zero on the 1st leg. gen> and where will I get it I say there’s no time, and even with the session it’s crap :( gen> I'll probably give up soldering until next year :) gen> while the information needs to be collected I have no doubt that in a year there will be much less information about specs in Internet. For example, MS IR23 was discontinued (deep body) about 10 years ago. gen> damn it, I was poking around with friends who work with computer maintenance gen> everyone assures: “that there are no programmers, but if necessary, there was something gen> to program it was given somewhere, and thank God there is now technology gen> is better and there is no need to flash anything ":( It’s as if in 90 there was no need for a programmer to build IBM, but they very good everyone needed it in 96(?) during the CIH virus epidemic. You always after the words “they gave it somewhere,” he could ask “where?”, “who knows?” and It’s more likely your reluctance to find anything. I specifically found the Pentagon-128 diagram on the Internet and downloaded it at terrible speed from people.ru: http://mr-linker.narod.ru/sheme_zx.html A heating chip in 90% of cases indicates a malfunction. What could cause the chip to heat up: due to the fact that it itself burned out and there was a short circuit in it something somewhere and for example if his legs 30 and 31 are locked to the ground it can be thought of as a regular resistor to ground, which will heat up if its other leg is closed to +5. Depending on resistance and power the resulting “resistor” may cause other chips to heat up or nothing at all warm up, of course, depending on the switching circuit. If your Pentagon matches the diagram from that site, then m/s D55 is the printer register and its outputs are not connected anywhere at all if there is no printer. so you can (should) safely unsolder/bite out this chip (see also recommendationsjtn on the first page of this topic) If you're lucky, the computer will work, but judging by the burnt-out RF2, this is not the case will happen. If you say that the location of the board did not make it possible to connect printer, you should think about the cause of the malfunction... A “desktop/tower type” case means a switching power supply, therefore ground at the power supply unit, regardless of whether it is turned on with a button or not half phase is present (~127V). Accordingly, when simply plugged into a socket computer if you connect something that also plugs into an outlet (for example TV) you can burn one of the devices. The second option is that the power supply has dried up its kanders and it produces “slightly” wrong voltage that should be output. Well, of course, there is always the option that the board touched the wiring or the board touched the case when it was turned on/turned on and this led to a malfunction. Judging by the fact that the stripes/chess pieces your Pentagon still gives chances to fix it yes. I have absolutely no idea about the contents of MS RF2, but there are suspicions that without You won't be able to run anything from the code on this chip. Perhaps there was a piece of glass lost in another incident and the chip is still working...

From Vasily Strochkin To All 16 December 2005

Hello,geners gen> yes, I know real people, sometimes I sometimes sometimes, but about Rybinsk gen> is generally a pipe :( gen> and so you tell me if the ROM failed, why couldn’t the diodes gen> the mikruha actually BURNED, even the window burst :) You wrote that the register, which turned out to be the printer register, was heating up? Don't go to a fortune teller - you connected the printer on the go...

From jtn To All 16 December 2005

Hello,geners then so 1. bite rf2yu (if you haven’t already) 2. using an oscillator, look at the scheme for the formation /CS for the fuse (27512) from A14,A15, MREQ and RD. If memory serves, RD is not used there - a glitch in the Pentagon circuit (well known) but the blocking signal is used when RF2 is turned on. printer key and ports are blocked by a signal on the printer port - rude speaking, when the printer cord is connected, there is a jumper in its connector that and turns on all this crap (you don’t really need it, but the jumper can be it is also soldered on the board, if the former owner was, for example, a fan of Covox). so, without a bug, with a mattress screen, on all the above lines (erome blocking) there must be impulses and as a result too - this moment is necessary track. 3. There are plenty of diagrams of the Pentagon everywhere, both in pseudographics and scanned form. 4. ega-card? what's that? everything has been discussed a thousand times a long time ago. p.s. Sorry if I'm rude =)