• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

AI troubleshooting

Status
Not open for further replies.
JJIrons69

JJIrons69

Emitter and Base Collector
Shoot - I'm also working on a 535 tuner. It sounds like they're broadcasting from the moon. At 3/4 volume it's still very quite. I'm right in the middle of testing everything to try to find the fault. Here's what Chat told me the other day - so YMMV. I haven't confirmed them all yet. Don't shoot the messenger.

AWE-046 terminal : Schematic label/function : What I'd expect with DMM
1 FM ANT RF input — no useful DC
2 FM ANT RF input — no useful DC
3 B+ ≈13–14 V DC
4 FE B Front-end connection
5 — front-end/switching connection
6 T. MET Tuning-meter output
7 MUT AUTO OUT Muting/auto control
8 T. MET Tuning-meter output
9 FM S MET FM signal-meter output
10 COMMON ≈0 V
11 FM L OUT audio output
12 E ground/common
13 FM R OUT audio output
14–17 switching/MPX circuitry state dependent
18–21 switching/control area state dependent
22 COMMON ≈0 V
23–24 AM-related connections
25 AM LOCAL AM circuit
26 AM S MET AM signal meter
27–29 AM/switching connections
30 Comes from the AM detector / signal-strength circuit through R53 (150 Ω) and the D1/D2 network AM S-METER output
31 Not used/shown on AWE-046 No terminal 31 on this board drawing
32 Directly tied to the ground/E bus Ground, ~0 VDC
33 Goes into the mode-control network around Q6/HA1156, at the node with R34 (1 MΩ) and C62 (0.001 µF) “TO +B EXCEPT ST” control input
 
Register to hide this ad
rcs16 said:
Well I guess I should be stepping aside, out to pasture l, chatgpt and AI can tell AK members how to troubleshoot their stereos. Good luck with that folks :-) Sad day, sad world it's has become.
Click to expand...
i was toying around recently and Ai was wrong 9 out of 10 times regarding phono stages. the one time it was right i told it that it was wrong and it agreed with me and apologized. :dunno:
 
Well folks, all good things come an end.
Starting to see more and more folks asking AI to help find faults. Even questioning or trying to verify my advise.
I guess I should be stepping aside, out to pasture so to speak, chatgpt and AI can take over and tell AK members now how to troubleshoot their stereos. Good luck with that folks :-)
Sad day the world has become imo.
So instead of advising maybe correcting members, I need to now educate AI as well, good luck with that. Isn't going to happen.
From what I've seen, AI bots scour the net/web for information without knowing if it's right or wrong. Imo a perpetual f'up machine in action.
So going forward if I see folks posting AI generated troubleshooting advise I will simply ignore those threads.
Have at it. :-)
 
Last edited:
rcs16 said:
Even questioning or trying to verify my advise.
Click to expand...
Not AI related but when I first started on Ak I asked for advice as I was trying to fix my Sx-990 and this guy i’d never heard of Markthefixer answered and I asked.. are you sure about that? Big mistake that was!
 
Last edited:
I ask my phone a lot of questions about things I’m curious about and I’m just now realizing how much misinformation is in AI answers. Most of my questions are about gardening issues. I’m not sure I would trust it for anything technical like stereo repair.
 
Well, there have been occasions that AI helped me solve a particular problem (don't have an example ready though) but I always have to do a lot of thinking and understanding myself afterwards. Actually I do have an example. I needed to write Python code to send text to an HD44780 display. AI (from DuckDuckGo) came up with very specific code but I ended up going a completely different route. But it did help me getting started. I have to admit...
 
Okay, so let the AI LLM tell you how to fix your broken stereo. All it will do is scour the net for relevant information not knowing what's right or wrong. Up load the schematic, mark up the voltages that you measure and ask it to analyze the circuit, perform simulations on how the circuit operates, AC and DC, determine what's defective, what parts to buy, what to substitute if the original part is no longer available,where to buy them. How to solder, how to test. Good luck with that, let me know how it goes in fixing your stereo.
 
The more AI slop I see here, the less I'm inclined to participate. I definitely think less of people who use it and I don't have a lot of respect for communities that tolerate, much less encourage it.
 
I don't use AI to troubleshoot anything ..... yet....But have used it for quite a few things and it worked very well. I gave it parameters for an excel worksheet and was done in seconds and work perfectly for my application. Assisted in some technical writing I was doing and completed it without a hitch in a few keystrokes. SO....it has its place.
 
Like always, I asked ai.

AI repair diagnostics right now is genuinely in that early stage. I can pull from service manuals, known failure patterns, forum threads, and general electronics principles — and sometimes that's actually useful as a starting point. But I haven't touched a Sansui. I've never smelled a blown cap or felt the resistance on a dirty pot or heard the difference between a cold solder joint and a failing output transistor.That tactile, experiential knowledge is something I'm genuinely years away from having any real equivalent of — if ever.
The AK crowd that gets loudest about AI "getting everything wrong" on diagnostics is usually reacting to someone who took an AI answer at face value and skipped the actual troubleshooting process. That's a people problem more than an AI problem — but it's a fair warning about the current limitations.
The honest version of what I'm good for right now is being a well-read research assistant who's never held a soldering iron. Useful in the right context. Not a replacement for 30 years of hands-on experience and a legendary ear. You'd know the difference in about 30 seconds.



Salesman tells you it’s perfect. Ai says it’s not and knows it’s not.
 
As seems to be the case with any thread having even the slightest connection with AI, the "AI GOOD" vs "AI BAD" debate has polluted this thread.

I've deleted the threadcrapping posts and will continue to delete off-topic posts (ie general AI musings, debates, predictions, etc).

Not sure why it's the same members who are compelled to debate this topic over and over and over and over and over, with the same statements, whenever AI arises, but do it elsewhere please.
 
Sorry guess I should of been clearer to make my post on topic. The design and troubleshooting AI we are required to use and develop at work is proving to be very good at both. :rockon:

Odds are the flaws of the current free offerings to the general public will be "trouble free" fairly soon.
 
Last edited:
Why is there no clear policy on AI threads? Why not have a place exclusively for those posts because they have nothing to do with audio. I feel like we are being forced to accept AI as a cure all to everything. And as was stated earlier then what is the point of learning actual troubleshooting.

There has not been a single post that responded to the actual content of OP's post. Other than someone saying that he might as well quit what he is doing because we can just let a robot figure it out.

Yes I am frustrated.
 
"There has not been a single post that responded to the actual content of OP's post."

Okay, fair response, how about try cleaning the switches. Probe the signal paths with a scope. Read the service manual, teach yourself something about electronics. Buy a DMM, it's a long list
 
Last edited:
GChief said:
We will have to agree to disagree on that :bigok:
Click to expand...
OK then. Tell me how that post relates to OP's post about his 535 tuner... The only post that had anything to do with it was from @rcs16 when he said that maybe it's time to step aside and let AI fix things, since human troubleshooting is no longer needed. Although I believe that he thinks that humans are still needed for the process and they will do a better job.
 
JJIrons69 said:
Shoot - I'm also working on a 535 tuner. It sounds like they're broadcasting from the moon. At 3/4 volume it's still very quite. I'm right in the middle of testing everything to try to find the fault. Here's what Chat told me the other day - so YMMV. I haven't confirmed them all yet. Don't shoot the messenger.

AWE-046 terminal : Schematic label/function : What I'd expect with DMM
1 FM ANT RF input — no useful DC
2 FM ANT RF input — no useful DC
3 B+ ≈13–14 V DC
4 FE B Front-end connection
5 — front-end/switching connection
6 T. MET Tuning-meter output
7 MUT AUTO OUT Muting/auto control
8 T. MET Tuning-meter output
9 FM S MET FM signal-meter output
10 COMMON ≈0 V
11 FM L OUT audio output
12 E ground/common
13 FM R OUT audio output
14–17 switching/MPX circuitry state dependent
18–21 switching/control area state dependent
22 COMMON ≈0 V
23–24 AM-related connections
25 AM LOCAL AM circuit
26 AM S MET AM signal meter
27–29 AM/switching connections
30 Comes from the AM detector / signal-strength circuit through R53 (150 Ω) and the D1/D2 network AM S-METER output
31 Not used/shown on AWE-046 No terminal 31 on this board drawing
32 Directly tied to the ground/E bus Ground, ~0 VDC
33 Goes into the mode-control network around Q6/HA1156, at the node with R34 (1 MΩ) and C62 (0.001 µF) “TO +B EXCEPT ST” control input
Click to expand...
It's unclear to me what steps were recommended and what you have done so far.
 
Status
Not open for further replies.
Back
Top Bottom