svofski has been working on a Compact Cassette data drive called "tapeshnik" for a while now...he started out with an off-the-shelf full logic mechanism but later designed and built his own.
Nette Bastelei :) Der Zähler scheint mir ein bisschen schnell zu laufen: er macht ca. 100 Schritte in 14 Sekunden, d.h. wenn er die ganze Zeit so schnell läuft, überdreht er nach ca. 23 Minuten. Wird er eventuell nach einger Zeit langsamer, weil er die Umdrehungen eines mechanischen Teils misst, das sich am Anfang der Kassette schneller dreht als am Ende der Kassette ? So war das beim Videorekorder meines Opas: am Anfang lief der Zähler schnell, sodass man dachte er wird überdrehen, aber dann wurde er immer langsamer und zählte ab 8000 nur noch sehr langsam, sodass er doch nicht überdrehte und das Band vorher zu Ende war. 11:21 C64 SID-Chip programmer took a hit of his bong (Music by Fanta) :) 16:07 lcd_string("SIMON"); -> Was wird denn da angezeigt ?
Also - ich hab nix gemacht! Von einem Compile auf den nächsten hat der Zähler plötzlich angefangen, die meisten Schritte doppelt zu zählen... ich hab nix an der Interrupt- oder Zählerlogik gemacht und mich nicht weiter damit befasst. Davor wars so, dass er bei ner C90 irgendwo bei 9800irgendwas geladet ist, also ne C100 oder C110 oder C120 würde überdrehen, aber jetzt überdreht er mehr als 2x pro Seite. Der Zähler hängt über Interrupt an der Lichtschranke vom Wickelteller, der auf Seite 1 das Band aufwickelt und auf Seite 2 abwickelt. Damit wird der Zähler langsamer, zählt aber immer noch so merkwürdig.
Mein 0D-PONG hat dasselbe Problem - Interrupts lösen mit oder ohne Debounce einfach doppelt aus. Ich muss einfach das Ergebnis durch 2 teilen, bevor ichs ans Display schicke (also 1xRechtsshift)
Hi! Nice!, I do have a question if you don't mind... Both of your open collector driver ICs are capable of handling 30v so why did you use the PNP transistors? I think they are not strictly necessary. Is it because you wanted to invert the signal from the uC?
I wanted to drive the power, not drive the absence of power by creating a driven short to ground and needing lower-resistance pull-ups instead of high resistace (100k, could probably work with 1m) pull-downs. NPNs (which is what the output stage of an open collector is IIRC) are good for driving to ground, PNPs are good for driving to Vcc. At least that's what I recall if I understood it correctly. This point is moot though, because it turned out 20V is plenty enough for a bright display - the CMOS 4000 parts can handle 20V. The banks of transistors were absolutely unnecessary.
@@senilyDeluxe I might be missing something, but I think t he value of the PNP pull down resistor and the NPN pull up resistor would be about the same. I think the real issue is quiescent power draw. If you're powering half of the display most of the time, then it doesn't matter, but if you're powering a fraction of it, then the PNP solution becomes more efficient.
@@Enigma758 I didn't study Electrical Engineering. I almost did - then I encountered a retro computer that needs to be programmed to work and suddenly I was into programming. So no clue if what I'm saying is right or what you're saying is right. The NSM pinball machines (rare stuff...) actually use a bank of NPNs to short the display to ground to turn the segments off, just as you suggest (and who am I to argue with a company making pinball machines?). But I'm thinking that's like having your light always switched on and just slide a board in front of it if you want it dark. The pull-down resistor can be fairly high resistance, because when the transistor is switched off, the output has a tendency to slightly float up. But if you use pull-ups, they need to actually pass current (even if it's only a few mA). The pull-downs are just for the stray capacitance and the tendency to float up. If you leave the pull-down resistor out, a dark segment will either stay dark or be glowing very dim, but once you switch it on, it'll stay lit (or light up a little less intense) if you switch if off. And it might take minutes to dim down. I want my displays to use more power if they light up and less power if they are dark.
@@senilyDeluxe Makes sense, I wasn't sure if the pull down requirements were less for PNP and along with your explanation, I see that you are using a 100k resistor for that. I agree, there are good reasons for driving high to power a device. Thanks for the feedback and I really enjoyed the video1
Bravo 👍 If i had that knowledge, i can repair everything.
Jupitec schleichwerbung und so^^
Schön gemacht, Simon!
svofski has been working on a Compact Cassette data drive called "tapeshnik" for a while now...he started out with an off-the-shelf full logic mechanism but later designed and built his own.
I wish I was! But I will return to it eventually.
@@svofski I've glanced over it. Looks amazing! Gonna watch these videos soon.
@@senilyDeluxe youtube is smaller than it initially seems ;)
Nette Bastelei :) Der Zähler scheint mir ein bisschen schnell zu laufen: er macht ca. 100 Schritte in 14 Sekunden, d.h. wenn er die ganze Zeit so schnell läuft, überdreht er nach ca. 23 Minuten. Wird er eventuell nach einger Zeit langsamer, weil er die Umdrehungen eines mechanischen Teils misst, das sich am Anfang der Kassette schneller dreht als am Ende der Kassette ? So war das beim Videorekorder meines Opas: am Anfang lief der Zähler schnell, sodass man dachte er wird überdrehen, aber dann wurde er immer langsamer und zählte ab 8000 nur noch sehr langsam, sodass er doch nicht überdrehte und das Band vorher zu Ende war.
11:21 C64 SID-Chip programmer took a hit of his bong (Music by Fanta) :)
16:07 lcd_string("SIMON"); -> Was wird denn da angezeigt ?
Also - ich hab nix gemacht! Von einem Compile auf den nächsten hat der Zähler plötzlich angefangen, die meisten Schritte doppelt zu zählen... ich hab nix an der Interrupt- oder Zählerlogik gemacht und mich nicht weiter damit befasst. Davor wars so, dass er bei ner C90 irgendwo bei 9800irgendwas geladet ist, also ne C100 oder C110 oder C120 würde überdrehen, aber jetzt überdreht er mehr als 2x pro Seite.
Der Zähler hängt über Interrupt an der Lichtschranke vom Wickelteller, der auf Seite 1 das Band aufwickelt und auf Seite 2 abwickelt. Damit wird der Zähler langsamer, zählt aber immer noch so merkwürdig.
achso lcd_string("SIMON"); - das hab ich nur aus Debuggründen eingebaut. Das kommt, wenn das Laufwerk in einem ungültigen Zustand ist.
Mein 0D-PONG hat dasselbe Problem - Interrupts lösen mit oder ohne Debounce einfach doppelt aus. Ich muss einfach das Ergebnis durch 2 teilen, bevor ichs ans Display schicke (also 1xRechtsshift)
Hi! Nice!, I do have a question if you don't mind...
Both of your open collector driver ICs are capable of handling 30v so why did you use the PNP transistors? I think they are not strictly necessary. Is it because you wanted to invert the signal from the uC?
I wanted to drive the power, not drive the absence of power by creating a driven short to ground and needing lower-resistance pull-ups instead of high resistace (100k, could probably work with 1m) pull-downs.
NPNs (which is what the output stage of an open collector is IIRC) are good for driving to ground, PNPs are good for driving to Vcc. At least that's what I recall if I understood it correctly.
This point is moot though, because it turned out 20V is plenty enough for a bright display - the CMOS 4000 parts can handle 20V. The banks of transistors were absolutely unnecessary.
@@senilyDeluxe I might be missing something, but I think t he value of the PNP pull down resistor and the NPN pull up resistor would be about the same. I think the real issue is quiescent power draw. If you're powering half of the display most of the time, then it doesn't matter, but if you're powering a fraction of it, then the PNP solution becomes more efficient.
@@Enigma758 I didn't study Electrical Engineering. I almost did - then I encountered a retro computer that needs to be programmed to work and suddenly I was into programming.
So no clue if what I'm saying is right or what you're saying is right.
The NSM pinball machines (rare stuff...) actually use a bank of NPNs to short the display to ground to turn the segments off, just as you suggest (and who am I to argue with a company making pinball machines?).
But I'm thinking that's like having your light always switched on and just slide a board in front of it if you want it dark.
The pull-down resistor can be fairly high resistance, because when the transistor is switched off, the output has a tendency to slightly float up. But if you use pull-ups, they need to actually pass current (even if it's only a few mA). The pull-downs are just for the stray capacitance and the tendency to float up. If you leave the pull-down resistor out, a dark segment will either stay dark or be glowing very dim, but once you switch it on, it'll stay lit (or light up a little less intense) if you switch if off. And it might take minutes to dim down.
I want my displays to use more power if they light up and less power if they are dark.
@@senilyDeluxe Makes sense, I wasn't sure if the pull down requirements were less for PNP and along with your explanation, I see that you are using a 100k resistor for that. I agree, there are good reasons for driving high to power a device. Thanks for the feedback and I really enjoyed the video1
I "found" this radio. Lol.
i don't understand a single word but i wish i did
What car is this cassette deck from?
A Smart.