glitch

Greetings, O venerable Viktor! [Monday March 13, 2000] at [00:00:00] Viktor Vorobov wrote a letter to All. They discussed the topic of "glitch". Let's continue what we started... VV> Settled in my (and has been for a long time) computer (yellow SCORP) subj. VV> Namely: when AY is connected (real AY-3-8912) not VV> work VV> some programs. VV> (started after some time VV> are reset/freeze). VV> It’s interesting that if you remove AY, then everything works fine. VV> If anyone has encountered something similar, then tell me WHAT to do. 1. The signal /M1 of the processor to the AY decoder is not connected/broken. AY shits on bus in the interrupt acknowledge cycle. The test program was a waste. 2. The computer bus may be overloaded. 3. Within small limits (+/- 0.1-0.2V) play with +5V voltage. Well, I found something else ancient, handwritten. :) I still have it done this way (at I'm not a Scorpio!). ================== begin of file "AY&ZXBUS.C" ================== Area : HARDWARE.ZX From : Felix Knajzev To: All Subj: AY “presses” the tire... Attr: 0 _Here_ Greetings, O venerable All! Even at the dawn of the Great Upgrade, I noticed something strangeinfluence of AY-3-8912A on the ZX data bus, even if in decryption the /M1 signal was turned on. It all came down to the fact that after installation of this MS in the computer, it (the computer) started out very glitchy work: when testing memory using TEST ROM, the RAM was working _very_ unstable; Every now and then errors appeared, and, moreover, failures occurred in different cells and different bits, i.e. full RND. The operation of the RAM was very much tied to the supply voltage (+5B) comp. I had to set it with an accuracy of +/- 0.1B (!). The CLK Z80 line, naturally, was pulled up to +5B by a resistor from 470 Ohm to 1kOm. I observed such glitching, at least, on four machines (and mine) where AY-3-8912A was installed! The last time I encountered this problem was in the spring of this year. when I upgraded my nephew's car ;). Those. during installation AY in the socket, the computer started to glitch “in black”. :( Speech We are talking only about AY-3-8912A chips made in Taiwan (of the goats! :). Dan- I have no information about the effect of AY-3-8910 on the tire. I got tired of it :) and it was decided to untie the tire computer data (Z80) and AY using a buffer which would have "quality" Z-state at the outputs is guaranteed. The choice has fallen to the “replaced one” :) KP1533(555)AP6. The guess turned out to be correct! After appropriate switching of signals for AP6, and installationthis "in the gap" between the AY-com and the Z80, the computer began to work also on- reliably (glitch-free), just like without AY! I could drive for hours TEST ROM and not a single failure! "Window" of the computer's operating voltages too expanded! How does AY load the bus? I can’t give an exact answer yet. There may be two reasons for this: a strong capacitive load, or that most likely, the AY data bus _inside_ the MS itself is tightened resistors to +5V. I just remembered that according to this principle Bidirectional parallel ports were built in this LSI. Check You can easily drink: you need to supply power to AY, transfer it bus into the Z-state and measure the current flowing to _ground_. If ta- which will take place - the answer is clear, otherwise... 2All: Maybe someone will try to check this? Or maybe someone already knows exactly how they were built AY inputs/outputs connected to the processor? If you experience similar glitches on your computer, I described above, then several circuit solutions will be given below, allowing you to significantly deactivate your car. ;) Buffer 1533(555)AP6 is connected “in the gap” of the AY data bus and the processor is always the same, only the feeding pattern will change signal to input /OE buffer: KP1533AP6 ┌──┬──┬──┐ ┌ D0>──2┤A0│BF│B0├18──>D0 ┐ │ D1>──3┤A1│<>│B1├17──>D1 ││ D2>──4┤A2│ │B2├16──>D2 │ To data bus Z80< D3>──5┤A3│ │B3├15──>D3 > data bus to AY │ D4>──6┤A4│ │B4├14──>D4 │ │ D5>──7┤A5│ │B5├13──>D5 │ │ D6>──8┤A6│ │B6├12──>D6 │ └ D7>──9┤A7│ │B7├11──>D7 ┘ __ ├──┤ ├──┤ RD>──1┤T │ │+U├20─o+5B ├──┤ ├──┤ Buffer sample signal>─19oOE│ │0V├10─┐ See below. └──┴──┴──┘ ▀▀▀ 1. “Old” AY switching circuit (like mine): LE1 ┌──┐ ┌──┐ A14>──┤1 o───2┤1 o1────>AY_BC1 └──┘ ┌─3┤ │ │ └──┘ WR_7FFD>───o───────────> to the 19th leg of the buffer. (see above) + /M1 │ LE1 │ ┌──┐ __ └─5┤1 o4────>AY_BDIR WR>─────────6┤ │ └──┘ 2. For “yellow” Scorpio. Option #1: 1/3 555LI3 (additional valve) D53 555ID7 ┌──┐ ┌──┬──┬──┐ ┌─────1┤& ├12───>on the 19th leg of the buffer. (see above) ─1┤A0│DC│0 o─15─ │ ┌───2┤ │ ─2┤A1│ │1 o─14─ │ │ ┌13┤ │ ─3┤A2│ │2 o─13─ │ │ │ └──┘ ├──┤ │3 o─12─ │ │ │ ─4oE0│ │4 o─11──┼─┼─o D50.1 ─5oE1│ │5 o─10─ │ │ │ ┌──┐ ─6┤E2│ │6 o─9─o─┘ │ └─1┤& o3────>AY_BDIR │ │ │7 o─7┐│ │ ┌─2┤ │ └──┴──┴──┘ ││ │ │ └──┘││ │ │ D50.2 ││ │ │ ┌──┐ │└───┼─o─4┤& o6────>AY_BC1 └────o───5┤ │ └──┘ 3. For “yellow” Scorpio. Option #2: D53 555ID7 ┌──┬──┬──┐ ─1┤A0│DC│0 o─15─ ─2┤A1│ │1 o─14─ D50.1 ─3┤A2│ │2 o─13─ ┌──┐ ├──┤ │3 o─12─ ┌────1┤& o3──o─>AY_BDIR ─4oE0│ │4 o─11───┘ ┌─2┤ │ │ ─5oE1│ │5 o─10─ │ └──┘ │ ┌──┐ ─6┤E2│ │6 o─9─┐ │ └2┤1 o1───>on the 19th leg of the buffer. (see above) │ │ │7 o─7┐│ │ ┌3┤ │ └──┴──┴──┘ ││ │ │ └──┘ ││ │ D50.2 │ 1/4 555LE1 (additional ││ │ ┌──┐ │ valve) │└─────o─4┤& o6──o─>AY_BC1 └────────5┤ │ └──┘ That's all, actually. If by any chance anyone has the data "cartoons" will be implemented on your computer, then I'll be happy to listen ;) Your "reviews"! Best regards, Felix! -+- iS-EDIT 5.05+ + Origin: CONNECT 2400/NONE (@addr) ==== end of file "AY&ZXBUS.C" ================== Best regards, Felix. [I.ZX]