|
ColdPie posted:Cool thread hope you fix it Thanks, but don't put any money on it!
|
#
?
Nov 19, 2021 01:33
|
|
|
|
| # ? May 3, 2024 20:05 |
|
|
https://www.researchgate.net/post/What-is-capacitor-short-circuit-fault I'm super curious now. after you pull c3, please please please measure the resistance across its leads
|
|
#
?
Nov 19, 2021 01:34
|
|
|
[quote="crusty" post="519362776"] Cool av, I did a GIS but couldn't see it. What's it from? LOL one of the mods gave everybody an AV from pulp scifi covers in the Trump thread a couple years ago. Other posters got nubile women in space shorts while I got this.
|
|
#
?
Nov 19, 2021 03:09
|
|
|
The Voice of Labor posted:I dunno, parsing out that a serial dac is being run through sample and hold circuits to get stereo channels while never having thought about it before is a pretty tough piece of inferior smartness to defend Lol, well I've been unemployed, so I've had a lot of thinking time. The Voice of Labor posted:I bet you're such a nerd you're going to look at the clock frequency and the demux/ s/h circuits and figure out what the effective sample rate is. and I bet you're going to use math to do it. then you're gonna figure out how the reverb circuit works and clone it into a stompbox WELL... I admit I thought about it.  I did some quick calcs in my head and got an answer that was different from what I'd already read somewhere (32kHz) although that might just be the rate of the stored samples and not the effective output rate - and decided I'd come back to it later. So maybe a lazy nerd.The Voice of Labor posted:telling you though Yeah, I'm going to spring for that. It will probably be a little while until I'm soldering parts back in - my new job starts in a week so I'll be a little happier then about spending money replacing something I already have.
|
|
#
?
Nov 19, 2021 12:04
|
|
|
petit choux posted:LOL one of the mods gave everybody an AV from pulp scifi covers in the Trump thread a couple years ago. Other posters got nubile women in space shorts while I got this. Ha, awesome! I think you probably did alright. "The tiny little dresses on the space girls..."
|
|
#
?
Nov 19, 2021 12:14
|
|
|
Temp post so I don't lose it - heading away over the weekend. PARTS Capacitors Looking at those big dead/suspicious capacitors first. Near the power jack (C3, C6) and also one on the other side that looks fine (C85):   SANYO 1000μF 16V CE 85° C D.E.9N HW I couldn't find much about "DE9N", but since the service manual just lists μF/V specs, maybe it doesn't matter. Near the output jacks (C42, C43):  1000μF 16V NX-IV (M)85° C 9 (2) DH From the CM-64 service manual:  The numbers for each of these looks to be just a Roland ordering number, not any original manufacturer part number. So I guess we go by the basic details there (1000μF 16V) and maybe some more clues on the caps themselves.  CM-64 manual warning/note on the "NX-4" capacitors:Use only above capacitors for replacement, or the tone color might be changed delicately Is there some special characteristic of a "NX-4" cap (NX-IV on the cap itself), or are they saying to use exact capacitance? If it was just a capacitance warning, why even are these NX-4 caps being used when the others have the same μF/V specs? Interestingly, the MT-32 service manual doesn't list any "NX-4" capacitors, so I wonder if the tone differs slightly between them? I've not heard a direct MT-32/CM-32L A/B test but maybe I can find one. I'll have to ask around to find more info on this, but hopefully I can avoid replacing any of the green NX-4 caps. Will cross-post the question a few places if no answer in this thread. Interesting info on these exact caps in someone else's unit I found while looking for info: https://www.vogons.org/viewtopic.php?t=71067 What looks like some "rust spots" near the caps could just be discolored glue. So hopefully those might clean up easily and I won't have to replace all of them. Diodes Some handy info on testing diodes: How to Test Diodes with a Digital Multimeter OK, so diode testing is a voltage drop. Standard drops based on diode material. A note from that link: In some cases it may be necessary to remove one end of the diode from the circuit in order to test the diode. This was a worry of mine - these components could have multiple paths between the 2 ends which I assume could affect testing results. A quick search shows that the side of the diode with the stripe is -ve (cathode) end. Also that when shown with this symbol: —▷|—, the | indicates -ve end. A look at the circuit diagram:  Confirms the stripe and symbol match. Small victories... Anyway, Let's see what we can see with testing in place: D1: 0 shown either way (not 'OL', just 0). Zero drop - assume this is dead (short). It's connected to one end of the dead cap C3, so that makes sense. D2: 0.26v measured both ways. Symbol is a little weird, found this info on Zener diodes... and CM-64 diode list shows Zener on one item - nice. D2 seems pretty clean, may not be damaged. I finally found this component in the MUTING area in the circuit diagram, but based on the apparent usual usage of Zeners (voltage regulation), not sure what it does there. I found a video that seems to indicate Zener diode show OL (inf) voltage drop measured in reverse (unless voltage exceeds the designated voltage value for the diode), so more digging required there. D3: .58v / 1.94v. No idea what this result means (2v seems big) but it looks unaffected. Part of the MIDI input circuit. tbc... crusty fucked around with this message at 13:17 on Nov 19, 2021 |
|
#
?
Nov 19, 2021 13:13
|
|
|
crusty posted:Temp post so I don't lose it - heading away over the weekend. Thanks for doing this thread, I really appreciate somebody demonstrating the entire process of diagnosing etc. ... I would have replaced that cap with any old electrolytic of the right capacity and tried turning it on already. Which you may end up doing, but it really helps me to watch you do this as thoroughly as you are. My AV is one of the best of the scifi cover art AVs, no doubt. I really like it.
|
|
#
?
Nov 19, 2021 15:33
|
|
|
The Voice of Labor posted:get everything over spec (ie, get 25 or 50 volt ones for the 16v all electrolytic caps of the same value have the same electrical characteristics. material type is the only thing other than capacitance and voltage rating that matters. e: if the service manual is telling you to use a particular manufacturer's caps, do it. the only time I've ever noticed an audible difference for different caps with the same capacitance was on filter circuits, like, that's an exception way more than a rule e2: I uhh don't think you're going to be able to find those. you could put in sockets, then you could swap out caps without soldering them and test a bunch of brands and find the one you think sounds best resistors have a resistance, a tolerance and a maximum wattage. there are some outliers like temperature variable resistors and that weird fused one on your board, but they're their own separate thing. I...I don't think I've ever bought an inductor, but they would come in different henrys quote:A note from that link: In some cases it may be necessary to remove one end of the diode from the circuit in order to test the diode. The Voice of Labor fucked around with this message at 01:54 on Nov 20, 2021 |
|
#
?
Nov 20, 2021 01:29
|
|
|
|
| # ? May 3, 2024 20:05 |
|
|
crusty posted:I found a video that seems to indicate Zener diode show OL (inf) voltage drop measured in reverse (unless voltage exceeds the designated voltage value for the diode), so more digging required there. Correct! Zener diodes, when you go past the breakdown voltage threshold in reverse, will basically dump current to pull the voltage down to the breakdown voltage. This is a super interesting thread. If you get frustrated, i'm a pretty competent repair guy (focus on ham radio gear but i can do audio poo poo too) with a good bench with good meters/oscilloscopes and a signal generator. I'll be happy to spend an hour or two on it to try to diagnose if you wanna pay round trip cheapshipping to Denver. yospos will vouch for me that i'm not gonna gank your poo poo, i have enough of my own. IF IT WERE ME: I would separate this into two parts: "the part that makes the noises" and "the part that amplifies the noises." The lovely electrolytic cap situation indicates to me that this mayyyy be an issue with the latter, as that section always requires more power and thus more filter capacitance on the power supply rails than the synth side. There's got to be a mono or stereo very simple signal path between the synth/wavetable chips and whatever amps they use to bump those to line level. I would break that connection (usually there are either resistors or capacitors connecting these two components) and test each individually; meaning i would play notes on a keyboard or whatever and listen for a low level output from the out pins on the synth chips, and then i would inject a 1khz sine wave into the amp inputs and look for line level out. Oh also i'd scope the DC rails. The fact that we're dealing with explodeycaps means that there's a non zero chance that the DC rails are dirty and have AC of god-knows-how-much warble on it, and that sort of thing tends to lock up audio amplifiers in oscillation like a 16 year old on prom night. The oscillation can often be outside of hearable range so just a speaker check isn't sufficient. Starring this one and will be following along!
|
|
#
?
Nov 20, 2021 22:30
|
|
