MiSTer FPGA Forum

Looks like the perfect thread for me, i have exactly the opposite problem.

I can get the MiSTer to tape load using an app from my phone. It works perfectly. I came across a Sony TCM-939 at a Thrift store for CAD$10 and picked it up to give it a try.

Issue i get seems to be the coupling as stated. From the phone, on the ADC tester the line comes in somewhere near the left side of the screen. With the App basically blasting full volume i get a trace that extends roughly to the middle. (This is on the default settings which i think is the 3.3V range rather than line in voltage)

When plug in the Sony, I instantly see the trace drag off screen to the left. When i press play Ill see it do the opposite dragging over to the right. The odd part is that i can adjust the volume and seemingly get the trace to do its crossings across the whole 3.3V range, but when i go into the spectrum core, things basically fail to work... I get a lead in loading tone that sounds a bit wrong... the bands are not equal in length and basically it just fails.

Id assume that this is the same issue. Its likely the same issue due to my running of the Sony from battery which would leave it to be floating... Its probably a good idea I assume to try to ground the sony to the MiSTer for a start

Eros wrote: ↑Mon Jul 18, 2022 1:51 pm Looks like the perfect thread for me, i have exactly the opposite problem.

Sounds like my experience too: viewtopic.php?p=51081#p51081

Did you try the 1k resistor solution? im thinking of opening up the tape recorder and seeing if i can mod it

Nah, I'm not really that interested in this solution because I can load tapes fine on my real ZX Spectrums. In any case, I thought the resistor should go into the cable somewhere (so no need to mod the deck)? Pgimeno is the person to ask, unless you know your way around electronics well - for me it's all black magic.

I did. It's working for me.

Yes it's in the cable, not in the unit. It's placed *between* both wires. I use that cable to load from the TZXduino and it works as intended.

The Spectrum core expects voltages that cross zero; that's very reasonable for audio, and how most retrocomputers managed tape input. The ADC can only measure positive voltages, not negative ones, so the Spectrum core has a threshold very close to zero, and whatever is under that threshold is assumed to be negative. It's not a bad assumption, given the limitations of the DAC, but it turns out to be bad with the MiSTer circuitry, which causes a weak shift towards 1.8V or so. The resistor across the wires helps it reach zero with ease and be the symmetric input that the core expects.
.

Yeah just a note that my parts came today and I also made a cable with basically 1K between line and ground. This changed it from not working to... working which was quite satisfying