Skip to content

Site map) Home > Getting Started > Building > Durango·X > Troubleshooting


Troubleshooting guide

Since the Durango-X computer is based on off-the-shelf components, it's a rather complex circuit. On the other hand, every needed function can be clearly located into specific components, A gazillion things can go wrong, of course, but with some method and a few observations, any possible fault may be easily figured out. A multimeter is essential for basic probing; an oscilloscope (nothing especially fancy) will be helpful, but not absolutely necessary. A logic probe can be quite useful, too.

Before any further investigation, a visual inspection may be helpful. Look for:

  • Short-circuits on the board -- including from pads to nearby vias.
  • Bad solder joints.
  • Bent IC pins -- try reseating any suspect IC.

Tip

For better understanding of this document, you may want to have a look at the Theory of Operation, although it's not really needed for troubleshooting.

v2 changes

These instructions are intended for the v1 issue of Durango·X. For the v2, replace all references to U8 (74HC132), using U32 instead (74HC245) and follow any particular notes added.

No signs of life (no LEDs nor picture)

Check power supply and power switch, if fitted. Measuring the voltage across the power pins of any IC (usually 14/7, 16/8, 20/10 or 28/14 for +/-) should ideally read out between 4.75 and 5.25 volts, although it should be fine with voltages as low as 4.5 V (prototypes have been running OK at a mere 4.25 V, even less!).

If measured voltage is way too low, and the power supply is trusted, check resistance between the power pins (being careful with your meter's polarity!). Because of capacitors charging up, you won't get an instant reading; but it should asymptotically lead to around 9-10 kOhm. Any unusually low value might be caused, in rare cases, by C7 or C6.

Power LED lights up, but no sound or picture

  • If an oscilloscope or a very good multimeter is available, check for a ~24.5 MHz signal at U15 pin 10 (74HC4040). If no good signal there: suspect X1.
  • If there is, check for a somewhat asymmetric ~15.7 kHz signal at U15 pin 15. If not: suspect U15. If frequency is exactly 12 kHz: suspect U17 (74HC21) as well.
  • Check for a quite asymmetric ~50 Hz signal at U19 pin 12. If not: suspect U19 but if more like 30.6 Hz: suspect U17 too.
  • Check for ~15.7 kHz at U18 pin 6 and for ~50 Hz at U20 pin 6. Suspect these chips in case of failure. Those signals should show up (inverted) at U22 pins 11 and 14, respectively.
  • If an oscilloscope is available, look for a valid CSYNC signal at U23 pin 6 -- or just some activity, near (but not quite at) 5 volts.

v2 changes

The updated values are: - U15 pin 10: 28 MHz instead of 24.5. - U15 pin 15 (when wrong): ~13.67 kHz instead of 12.

If all of the above signals look OK, it might be a problem in the analogue section of the video outpuct circuit. Use a multimeter to check the following voltages (all referenced to GND). They do not need to match exactly the specified values and will certainly be dependent of the actual Power Supply voltage, but any deviation over ~10% is suspicious).

  • First of all, verify R19 (one pin will carry a valid 5 Vpp CSYNC signal while the other should stay around 1 volt.
  • 5 Volts between C4's pins; check this and the power supply otherwise.
  • 1.6 Volts at the base (centre pin) of Q4; if not, suspect R17, R18 or Q4.
  • 1 Volt at the emitter (right pin with the flat side towards you) of Q4; if not, and the above is correct, suspect Q4 or R21.
  • 3 - 3.7 Volts at the base (centre pin) of Q5 (actual value depends on overall picture brightness); if not, suspect R16 or Q5.This very same voltage should show up at the collector of Q4, leftmost pin, as they're interconnected.
  • 2.3 - 3 Volts at the emitter (right pin with the flat side towards you) of Q5 (actually ~0.7 V below the above one); if not, suspect Q5, R30 or C5 (and, if the second composite output is mounted, R31 and C9 as well).

Monitor detects valid signal, but screen is black

Note

Lack of image (but good sync signals) is likely to be a fault on some of the following items but, if nothing seems to fix the problem, check U2 SRAM and U6 as well.

Check logic level at U321 pin 11 for currently set video mode: "1" (~5 V) for HIRES, "0" for colour. If this pin lacks a stable value, suspect U428 or U321. Once the current video mode is determined, check the following:

HIRES mode

After checking R220 in the very first place, look for activity at the following pins, preferably in this order:

  • ~6.1 MHz signal at U16 pin 9 (suspect U15) and U224 pin 7 (suspect U16)
  • 767 kHz square wave at U16 pin 5 (suspect U15) and U321 pin 13 (suspect U16)
  • Same 767 kHz at U22 pin 9 (suspect U321)
  • U321 pin 3, like CSYNC but quite more symmetric, near 2.6 Volts (suspect U16, maybe U15 and/or U19)
  • U22 pin 4, like the above (suspect U321, maybe U428)
  • U227 pin 10, again like the previous signal (suspect U22)
  • 767 kHz asymmetric signal at U224 pin 15, maybe measuring near 4.7 Volts (suspect U227 or U15)
  • U23 pin 9 (suspect U224 or R824 if fitted)
  • U227 pin 9 (suspect U23 or RV231)
  • U23 pin 12 (suspect U227 or R825 if fitted) and pin 11 (U23 itself or R220 is to blame)

Colour mode

After checking R107-114 in the very first place, look for activity at the following pins:

  • ~1.5 MHz square wave at U124 pin 11 and U321 pin 14 (suspect U16 or even U15)
  • Same signal at U22 pin 9 (suspect U321)
  • U321 pin 5, like CSYNC but quite more symmetric, near 2.6 Volts (suspect U16, maybe U15 and/or U19)
  • U22 pin 5, like the above (suspect U321)
  • U127 pin 19, again like the previous signal (suspect U22 or RV128)
  • U124 pins 12-19 should show activity in some of them, unless a completely blank screen is on display -- powering up without cartridge should create a random image in case of doubt, adequate for this test.
  • Similar but somewhat faster activity at U126 pins 1, 4, 9 & 12 (suspect U125)
  • Pretty much the same at U127 pins 2-9 (suspect U126)
  • A similar kind of signal, but interrupted every ~64 µs, should show up at U127 pins 11-18.

Garbage is displayed, but no further activity

  • Check /RESET signal (U1 CPU pin 40), upon powerup should go low for a split second and then become (and stay) high.
    • If it doesn't, but the computer operates normally after pressing the RESET button: suspect C1, perhaps R3.
    • If it stays low all the time: suspect U8 (74HC132) and/or C1.
    • If RESET button makes no effect: suspect U8 (74HC132).
      • In any case, if while holding RESET the ERROR LED does NOT stay lit: suspect U8 (74HC132) or U12(74HC74).
  • If an oscilloscope or a good multimeter is available, check for a ~1.5 MHz signal at VCLK (U1 CPU pin 37). If no good signal there: suspect U16 (74HC02), perhaps U15 (74HC4040).
  • Try another cartridge and/or U2 SRAM (displayed pattern is likely to change between different SRAM pieces). Running without a valid cartridge may display similar symptoms.
  • U1 65C02 CPU might be dead, too. Look for an irregular signal at U1 CPU pin 7, around 450 kHz (will vary widely) and, if possible, activity at address pins (9-20 and 22-25), especially the lowest ones.
  • If of the above looks OK: suspect U9 (74HC00). Check U6 (74HC157) as well.

Erratic behaviour (LEDs flashing, garbage on screen, chirpy sounds...)

Suspect U10 (74HC139)

Software appears to start up, but no keyboard of gamepad effect at all. IRQ test fails.

Check ERROR LED. If lit: suspect U12 (74HC74). Otherwise suspect U8 (74HC132) or U14 (74HC4040)

Poor/no keyboard/gamepad response and/or erratic timing on some software

This is likely to be an interrupt problem.

Computer runs OK, but colour programs go instead into greyscale mode

If Durango-X cannot go into RGB colour mode, or doesn't stay reliably in it, first of all make certain that U321 pin 9 is at logic "1" (~5 V). If so, check Q306 (its emitter should measure at around 4.3 V) and R122. Some TVs may be quite picky about this resistor's value, thus you may want to try a lower value (say, 100 ohms).

Warning

In RGB mode, R122 accounts for almost one third of Durango's power consumption. Lowering the resistor's value will draw further power, thus make sure you're using a properly dimensioned Power Supply.

Marginal picture quality, with choppy pixels, ghost lines...

As the timing of the video circuitry is very tight, some adjustments have been provided. See Video Output Calibration for details.

Apparently running but only chunks of valid image visible, maybe repeated on

the screen; perhaps affecting just one video mode

  • Check U7 and/or U6. Depending on affected video mode, check also:
    • U104/105 for colour mode.
    • U204/205 for HIRES.

Tip

Swapping ICs between each pair might help detecting the malfunctioning chip. While U6/U7 are specified as 74HC157, you may use 74HC257 as well.

Picture OK, but distorted sound or no sound at all

  • Check U12 and, most importantly, Q3. Check D2, R4, R5, C2 and C3 as well.
  • Rarely, D7 might cause problems; you may remove it with no more effect than losing the audio feedback during Raspberry Pi transfers.
  • If the internal buzzer is being used: check R23 (and the buzzer itself). Weak power supplies may be troublesome while using the internal buzzer.

Note

Some cartridges may include custom sound hardware. If so, check D3 or try another sound-enabled cartridge.