Actions
Forum Thread Details
  • Replies 12 replies
  • Subscribers 301 subscribers
  • Views 1112 views
  • Users 0 members are here
Related
See a helpful answer?

Be sure to click 'more' and select 'suggest as answer'!

If you're the thread creator, be sure to click 'more' then 'Verify as Answer'!

MC1377P video circuit drops to almost 0 when terminated with 75 ohm. Any ideas?

baldengineer

A project I've been working on for several months, or years depending on how you count, is build a Mini Apple IIe. (Originally it was a portable, now its just a mini.) The key to the project is the MEGA-II ASIC from the Apple IIgs. It contains all of the support logic found in old Apple II computers. (Approximately 80 7400-series chips!).

 

Since no one has ever used the MEGA-II in this way, I decided to build a couple of prototype boards to test three key sections: 6502, MEGA-II, and VGC. Turns out, we have the 65c02 and MEGA-II to be seemingly working. It even attempts to boot the Apple IIe ROM. (And then it appears to panic.)

 

So, I moved our attention to the video board. There are two basic sections on the board: the digital side and the analog side. The digital side consists of the VGC (video graphics controller) ASIC and three 4-bit DACs for RGB. Long story short, the MEGA-II generates the classic Apple II video bitstream, the VGC+DAC section converts it to analog R,G,B.

 

The analog part of the video circuit is based on an MC1377 RGB-to-NTSC encoder. That is the part of the circuit giving me trouble. In this iteration of the project, I just copied the circuit from the Apple IIgs. (My final project won't have composite.) The problem is that when I attach a video receiver or terminate the MC1377's output with 75-ohms, its output drops from 2.6 Vpp to 300 mVpp. A bit too small.

 

Here's the circuit:

image

The Full KiCad file is available here.

 

Here's the comparison of output:

image

In both cases I am, effectively, probing at J18's signal. On the right, I terminate that node with a 75 ohm resistor. (Same behavior with an actual composite receiver.)

 

A bunch more waveforms and link to MC1377P datasheet here: https://www.baldengineer.com/mc1377-measurements-on-mini-apple-iie-prototype-video-board.html .

 

Any ideas why the output is clamping so hard?

Parents
  • Found It!

     

    There is a 220 ohm resistor on pin 9. pointed out that the datasheet says when used with a 150 ohm resistor, the MC1377 should be drawing 60 milliamps. Initially, I misread the passage and dismissed it because the current draw was around 40 milliamps.

     

    Realizing we should be drawing 60 milliamps, I rechecked that resistor. A bad solder joint had it floating. After touching up the solder, the MC1377 now draws 60.0 milliamps (strange, right?). Most importantly, the video signal looks correct loaded, AND unloaded.

     

    image

    bar pattern on cold boot

     

    image

    bar pattern on warm boot/reset

     

    Based on my previous logic analyzer traces, I'm pretty sure the ROM is getting stuck very early in the boot process. It finds its way into the interrupt service vector, despite having IRQ de-asserted with a pull-up. Makes me think the code gets lost at some point. (However, the first 10 or instructions look right when compared to the IIe's ROM file.)

     

    But now that video outputs something and appears to output valid characters, I can get back to the boot process.

  • in reply to baldengineer

    Unrelated question: What video capture hardware are you using to display the composite video from your awesome project in OBS?

    Thanks,

    - W.

  • in reply to wolfgangfriedrich

    Currently, I'm using a RetroTink 2X Pro to capture the composite signal. That goes into a cheap ($20) HDMI-to-USB adapter.

     

    The RetroTink does have a little bit of trouble doing color with the IIgs video signal. I haven't tested the color modes on my board yet. But I copied the IIgs video circuit, so I suspect I'll have the same issue.

     

    Interestingly, the IIgs and the IIe generate the composite signal very different from each other. The II/II+/IIe all do the funky Woz stuff, which the RetroTnk handles fine. From what I can tell, the IIgs circuit is way more "legal" NTSC. But the RetroTnk can't lock on to the color burst for some reason.

  • in reply to baldengineer

    Your device chain generates really good quality. I have cheapo composite video-2-USB stick (can't find the exact model) and it has horrible artifacts and also has a tough time to lock onto NTSC colour signal.

Reply Children
No Data