|
Post by Peer on Dec 28, 2022 13:45:24 GMT -5
I recently tried to "upgrade" my Vectrex console with a new DAC. If you want to know why, see here, but this is not required for what I am going to write / ask here. The original DAC inside the Vectrex is a Motorola MC1408P8. According to my research, that chip was discontinued by Motorola and licensed to TI and henceforth produced as DAC0808LCN/NOPB. From what I read it is said to be 100% technologically identical and compatible to the MC1408P8. I ordered 5 new DAC0808 from Mouser and experimented with them. I am having a peculiar problem with all of them, when I use them inside my consoles, and I do not understand what the reason could be, and if this is due to some in spite of everything existing incompatibility, or if possibly all the DACs I have received have some production flaw. Maybe someone with a better understanding of electronics could chime in? First of all, my two consoles are in good shape, and the two original MC140808 work without any problems. When I replace an original DAC with a new one, at first, everything works fine and the console runs without any problems. Switching the console on and off, unplugging, plugging in again, swapping cartridges, everything works fine. But when I leave the console switched off and with the power plug removed from the power outlet for a longer time (several hours), and then plug again and switch on the console again, it does not work anymore. This is best described as "the console behaves just as if the DAC was completely removed from its socket". As if the chip or some other part of the console was in some non-working state. The only remedy I found: remove the DAC from the socket. Switch the console on and off without a chip in the DAC socket, and then plug in the DAC again. After that, the console works fine again. Of course this is not an acceptable workaround. I must have swapped the DACs some hundred times. This problem does not occur with the original DACs, but with all the 5 new ones. I checked the datasheets, but with my layman’s knowledge I did not spot any significant difference. Does anybody out there have an idea what could be the cause for this? And has anyone else out there ever tried a replacement DAC in their console? Any help is appreciated.
Many Cheers, Peer
|
|
|
Post by gauze on Dec 28, 2022 13:52:03 GMT -5
wow that makes absolutely no sense, good luck
|
|
|
Post by kokovec on Dec 29, 2022 12:19:57 GMT -5
Were the original chips socketed? Could be an oxidation issue on the old sockets. I usually replace the sockets if I remove a chip. Sometimes resoldering the old ones does the trick. If not, crack on the PCB board? I always check for mechanical issues before I move on to more complex testing.
|
|
|
Post by Peer on Dec 29, 2022 12:53:17 GMT -5
Were the original chips socketed? Could be an oxidation issue on the old sockets. I usually replace the sockets if I remove a chip. Sometimes resoldering the old ones does the trick. If not, crack on the PCB board? I always check for mechanical issues before I move on to more complex testing. Thanks for your feedback. Yes, there is a socket on the board (and I have cleaned it some time ago). The weird thing is that the original DAC works perfectly when inserted in said socket, but the new DACs show the behavior described in the initial post when inserted in the very same socket. And that happens on both my consoles. That is why I think it highly unlikely that the problem is caused by either the socket(s) or some crack on the PCB board(s).
Right now, the original DACs are in place again on both my consoles, and they have been running fine without any issues over the past two days.
My only logical conclusion is that there must be some essential difference between the old and the new DACs. I am planning to run the same tests on my third console, which is at the university lab. If the behavior is the same, then the issue is definitely with the DACs and not the consoles.
But any theory or suggestion is very welcome!
|
|
|
Post by kokovec on Dec 30, 2022 18:40:07 GMT -5
I read Malban's blog post on the DAC (my head now hurts). I looked at the datasheets and they seem to be identical. That doesn't necessarily mean that they are. At this point I'd break out the oscilloscope and try to perform the "DAC Zero Test". If that doesn't work then there might be a current ramping time issue that's causing problems in the post-MUX circuitry. Everything downstream of the DAC is voodoo analog land.
|
|
|
Post by Peer on Dec 31, 2022 3:01:33 GMT -5
I read Malban's blog post on the DAC (my head now hurts). I looked at the datasheets and they seem to be identical. That doesn't necessarily mean that they are. At this point I'd break out the oscilloscope and try to perform the "DAC Zero Test". If that doesn't work then there might be a current ramping time issue that's causing problems in the post-MUX circuitry. Everything downstream of the DAC is voodoo analog land. Yeah, headaches are a commong side-effect of vectrexing
Additional information: I already did the "DAC Zero Test" with the new DACs when I inserted them. I also did the tests on my third console. Same phenomenon there. Either all the new DACs I received have a production flaw (improbable, but not impossible), or they are simply not 100% compatible with the original DACs (bad, as this would mean that there is no proper replacement). I have contacted Mouser and requested their help and possibly an exchange of the delivered DACs for new ones), but I have not yet received an answer.
Some "current ramping time issue that's causing problems in the post-MUX circuitry" sounds like a good trace. Especially as there is the "switch the console on and of with an ampty DAC socket" step which is necessary once the console/DAC has gone into the non-working state. But I have to admit that I have no clue how to proceed from here on.
"Voodo analog land" is a great term
|
|
|
Post by gauze on Jan 3, 2023 13:15:44 GMT -5
the only answer I got from electronics friends is something about "forward diode biasing on power up " but I am not sure they completely understood what I was saying.
|
|
|
Post by Peer on Jan 3, 2023 15:47:38 GMT -5
the only answer I got from electronics friends is something about "forward diode biasing on power up " but I am not sure they completely understood what I was saying. Thanks a lot for checking and getting back to me! I have to admit that I do not understand the details of this, and how this could be related to the new DACs, but I will try to follow this lead and investigate.
|
|
|
Post by playvectrex on Jan 4, 2023 10:39:40 GMT -5
Hi Peer!
I've been studying the datasheet and Vectrex logic board schematic to try to figure out why this might be happening, and I have a few ideas to try.
First question is when the new DAC doesn't work, do you still see a dot in the middle of the screen? I know you said it works like the DAC is missing, but just curious here.
If you need to switch on the unit with no DAC first, then plug in the DAC, it kind of sounds like the power rails need to be partially charged up a bit for it to work. Maybe the new DAC is presenting more of a load to +5V_ANA and -13V, and when they are completely discharged, that DAC load is preventing them from rising enough to remove the load?
If you measure +5V_ANA and -13V at the DAC when the DAC is in a non-working state, Vectrex powered on, we might get some insight about it. I would also measure the output of IC304 (LF353) pin 1 to see if anything is coming through voltage wise. -13V is generated with a voltage doubler circuit using capacitors and diodes, and as such it wouldn't take much load to fold it back to 0V, especially if you've never re-capped your board and those components are leaky. If we find out -13V is folding back, I would immediately re-cap the -13V supply area and test all of the diodes, and see if that fixes it. It's likely not folding back the +5V_ANA since that's powered with a fairly hefty linear regulator.
|
|
|
Post by Peer on Jan 4, 2023 10:45:21 GMT -5
Hi Brett,
thanks a lot, that sounds like a good lead. I will do the measurements as soon as I find the time and then report back.
Do I understand you correctly in that this phenomenon should not happen on a Vectrex with new (replaced) caps? I know someone who has new caps in their console, and I could ask them to try one of my new DACs in their machine.
Any idea / explanation, why this phenomenon does not happen with the old original DACs?
Many Cheers, Peer
|
|
|
Post by playvectrex on Jan 4, 2023 11:00:14 GMT -5
Sure thing! Right now it's just a hunch based on the behavior you describe. There's really not much other way to explain it in my mind... since the DAC is a pretty simple device. As long as it has power, and reference currents, it can convert the data bus values to current outputs. If the new DAC does have more of a start up load than the original, it could make it more susceptible to folding back the -13V rail which could cause other problems I don't know about. The datasheet says the Vee (pin 1) should be 4 volts below ground reference minimum... so I'm not sure what happens if that's not the case. Seems like a good enough plan to measure yours first... and try your friend's second. Looking forward to what you find
|
|
|
Post by Peer on Jan 6, 2023 10:51:32 GMT -5
... Looking forward to what you find Next time I will be at the lab is on Tuesday. I will try to do the measurements then.
Cheers, Peer
|
|
|
Post by Peer on Jan 10, 2023 13:05:58 GMT -5
... Looking forward to what you find And Hi again Brett!
Today I tried one of the new DACs in my lab vectrex:
The machine was cold (not switched on for several days) when I inserted the chip. I switched it on, and (at first) the display did not work. There was the startup tune and the Mine Storm sound, but the screen was blank. Only when I turned the brightness knob to max, a small dot (or rather very short line segment) could be seen at roughly the 2 o’clock position, some centimeters distance from the center.
I was just about to start measuring, when suddenly the soft humming of the circuitry changed and the vectors were there. It may have been some 50 seconds between switching the console on and the vectors to appear. Everything fine, from there on. Note that this console normally, i.e. with the original DAC, does not suffer from any warm-up symptoms and fires up very quickly, with just a second until the vectors appears. This new-DAC-at-first-not-working-but-the-after-some-delay-working has not happened this way on my home consoles, but I have probably not left them on long enough in the non-working state? I will try that later.
Anyway, from that point on, I could not bring the lab console back into the non-working state again, but only because I had to leave after an hour, so it was likely not switched-off long enough. I will be back at the lab tomorrow and then continue the experiment.
I still took the measurements you asked for while the new DAC was working. There was 13.078 Volts on pin 3, which according to my datasheet is the VEE pin. And there was 4.982 Volts on pin 13, which is the VCC pin. This probably does not tell much, as the DAC was working fine then?
Tomorrow, I will also bring one of my home consoles to the lab, and then try to get the measures on a non-working DAC. And I will have access to a third console which got new caps about 8 years ago. I guess, in vectrex years, this still counts as new 😉
Let me know if there are any additional measurements I should do.
Many Cheers, Peer
|
|
|
Post by kokovec on Jan 11, 2023 13:03:23 GMT -5
I wonder what would happen if you heated up the chip before turning on a cold Vectrex. Or hit it with some freeze spray once it's working.
|
|
|
Post by Peer on Jan 11, 2023 13:21:53 GMT -5
I wonder what would happen if you heated up the chip before turning on a cold Vectrex. Or hit it with some freeze spray once it's working. I will probably give that a try. But I think that this is not temperature-related, but rather something like all-capacitors-fully-depleted versus some-voltage-or-current-residue-still-present-somewhere. I found out that, after switching off and unplugging the console, there is still voltage present on pin 3 of the DAC socket. While operating, there are the proper -13 Volts. When switched off and unplugged, there are still some -1.5 Volts left on that pin, which are very very slowly drifting towards 0. Like really slow. After an hour it is still around -1.2 Volts. This residue must come from the power board. Remove the power board connection, and the pin goes immediately to 0. Connect again, and the residue voltage is back on that pin. This happens on all my 3 consoles. I can only guess, that this plays some role. As if the new DACs require some such residue voltage to fire up, but do not start on a "cold" console with no residue (0 Volts). And for some reason the old DACs do not need this. Pure speculation, don't know if that makes any sense....
|
|