EDIT: Thanks everyone for you help, that has been very instructive. I think I just have a very poor quality cable adapter. Given that Blueretro is mostly an opensource DIY project, I’ll make a cable adapter myself instead of trying to fix what would obviously not function properly.

Hi everyone,

First of all let me say that I’m a total noob in electronics (I really only know the basics) and I’m facing an issue that I really don’t know how to tackle.

I have bought a Blueretro NES adapter on Aliexpress (this one) and it does behave erratically when powered by the console alone (Bluetooth not working, LED indicator down, random outputs to the console).

When I’m powering via USB, everything function properly.

So I guessed that I might have a voltage issue on the NES side. I tested mine and make a few friend test theirs (5 in total including mine) and the result is still the same: the controller ports outputs between 4.6 and 4.8V instead of 5V.

The Blueretro itself apparently uses an AMS1117 (picture here) which, from my understanding, is stepping down 5V to 3.3V (wild guess, I don’t really know what it does, just quickly read the datasheet).

So, sorry for the long intro, here are my questions:

  • Is it wise to try to step up the voltage from the NES to the Blueretro from 4.6V to 5V? How would it be possible? Is it even possible?
  • Given that the Blueretro is taking 3.3V apparently, is it possible to step down from 4.6V to 3.3V instead? Is it wiser than stepping up?

Thanks in advance and sorry for the long post :)

  • @[email protected]English
    5•5 months ago

    I don’t have specific experience with game consoles, but the erratic behavior when powered by the console suggests that the port’s voltage is sagging when the Blueretro is attached, possibly lower than what the AMS1117 can tolerate.

    A quick search seems to show that the AMS1117 has a minimum dropout voltage of 1v. So for 3.3v output, the input must not drop below 4.3v. Other Low Dropout (LDO) regulators could have a smaller dropout voltage, but that might not be the root-cause.

    It’s possible that without load, the port provides 4.6-5v. But when loaded, it dips below 4.3v, producing the behavior you see. The problem then becomes: is it the NES that’s not providing sufficient current on the voltage bus, or is it the Blueretro trying to draw too much current?

    Are you able to measure the port’s voltage bus when the Blueretro is attached? That would help prove if the bus is sagging. Does the Blueretro allow you to use USB power when plugged into the console?

    • @[email protected]OPEnglish
      3•5 months ago

      What I’ve measured so far:

      • from the controller port directly: 4.8V
      • from the wires inside the adapter without Blueretro attached: 4.6V
      • from the wires inside the adapter with Blueretro attached: 3.9V

      There are some active components inside the adapter itself, I guess this could explain the drop 0.2V drop.

      So is it sagging?

      • @[email protected]English
        2•5 months ago

        The 74AHCT125N appears to be a 3v to 5v level shifter, which I wouldn’t expect to be drawing a lot of current. So if just the adapter’s presence is causing 0.2v drop, then yeah, that would suggest the output current of the port’s voltage rail isn’t very high, causing substantial sag.

        To be abundantly clear, is the Blueretro meant to work when only powered by the console? It certainly wouldn’t be as convenient – an extra cable and power block to deal with – but everything is functional when externally powered by USB, yes?

        Do I understand that your original post is debugging why the console isn’t sufficient to power the Blueretro?

        • @[email protected]OPEnglish
          2•5 months ago

          The 74AHCT125N appears to be a 3v to 5v level shifter

          I don’t think the cable adapter is using this particular chip. I see two much smaller ones. Maybe that’s the issue.

          To be abundantly clear, is the Blueretro meant to work when only powered by the console? It certainly wouldn’t be as convenient – an extra cable and power block to deal with – but everything is functional when externally powered by USB, yes?

          That’s right Blueretro is supposed to be powered by the console alone. When powerd by USB, it is perfectly functional.

          Do I understand that your original post is debugging why the console isn’t sufficient to power the Blueretro?

          That’s correct.

    • @[email protected]OPEnglish
      2•5 months ago

      I can respond to at least your second question right now: I can power the Blueretro from USB while on the console that’s how I know it is functioning properly when powered properly (my original post wasn’t clear enough I guess).

      I’ll test the voltage while powered on as soon as I can.

  • @scrionEnglish
    2•5 months ago

    Let me chime in with a few thoughts, can’t go into full detail since I’m currently traveling and on my phone:

    1. It should be possible to power the BlueRetro using the controller port only. IIRC, the controller pins providing power are directly connected to the internal NES regulator, so with enough headroom to spare, you should be good to go. Eventually, this will depend on the firmware of the ESP-WROOM module used by the BlueRetro (max. current draw for the ESP32 should be around 250mA with all peripherals enabled and transmitting, I’d expect the BlueRetro to use quite a bit less in normal operation). That’s also what the 8BitDo is doing, running at around 100mA.

    If you look at YouTube videos, you’ll also find one or two people running the BlueRetro directly off of the port.

    1. Where did you get the level shifter cable from?

    2. I’d also like to know the results of your voltage measurement in operation

    3. If push comes to shove and you decide to replace the LDO, the TLV1117LV made by TI would be a potential candidate, should be a drop-in replacement with better specs (price is about $0.30)

    • @[email protected]OPEnglish
      2•5 months ago
      It should be possible to power the BlueRetro using the controller port only. IIRC, the controller pins providing power are directly connected to the internal NES regulator, so with enough headroom to spare, you should be good to go. Eventually, this will depend on the firmware of the ESP-WROOM module used by the BlueRetro (max. current draw for the ESP32 should be around 250mA with all peripherals enabled and transmitting, I’d expect the BlueRetro to use quite a bit less in normal operation). That’s also what the 8BitDo is doing, running at around 100mA.

      That’s what I thought too. I originally ordered 4 cables (NES, SNES, N64 & GC). I only had troubles with the NES so I decided to just get a replacement. But the replacement has the exact same behaviour, so this is either an issue with my NES or the cable is not designed properly for PAL NES.

      Where did you get the level shifter cable from?

      What do you mean by that? The cable itself? It’s from Aliexpress.

      I’d also like to know the results of your voltage measurement in operation

      As stated below:

      • 4.8V on the controller port itself
      • 4.6V inside the cable adapter without Blueretro Core attached
      • 3.9V inside the cable adapter with Blueretro Core attached
      • 3.53V on input, 2.61V on the output of the AMS1117 when powered by the console only (and the Core itself is not functional that way)
      If push comes to shove and you decide to replace the LDO, the TLV1117LV made by TI would be a potential candidate, should be a drop-in replacement with better specs (price is about $0.30)

      Will it change something when the input power is so low? I feel like I might as well connect the 5V rail directly to the 3.3V input on the Blueretro…

      • @[email protected]English
        2•5 months ago

        If the controller port is connected to the same +5V rail as the CPU, wouldn’t the NES crash if it only got 4.6V or less?

        If only the BlueRetro is affected, maybe something behaves like a resistor in series with it, for example a broken solder joint in the adapter or at the connector on the NES

        • @[email protected]OPEnglish
          1•5 months ago

          Only Blueretro is affected. Maybe the Blueretro NES adapter cable has just poor quality components. The other cable adapters I have for other consoles work fine, including the SNES which also operates in 5V.

          • Dave.English
            2•5 months ago

            Could be. The way the voltage sags at the end of the cable with the blueretro connected suggests that someone skimped on copper in the wires.

            If you can figure out a way to measure the voltage at the port with the cable attached and powering the blueretro, do that. If it is mostly the same whether the blueretro is connected or not, yep, it’s the cable.

            • @[email protected]OPEnglish
              2•5 months ago

              I managed to do that (not an easy fit but I think I’ve achieved the measurement properly) and you’re right: the cable itself it probably of very poor quality. Given that Blueretro is mostly a DIY project I think I’ll make the cable adapter myself instead of trying to fix what is obviously not fixable.

              Thanks for the help!