[FIXED] Yamaha A-S700 intermittent power-on failure (and eventually full failure)

[Doodski]

I realise how important is 1) understanding the theory behind each type of component and
2) this is physical world, not just theoretical models, so physical properties and behaviour change with time, temperature, voltage, current flowing through, etc.
This and so much more are taken into account during design

I will try and make this fast because I have barber appointment at 11 am in ~35 minutes. When you see a proper textbook detailing the operations of capacitive reactance and inductive reactance and see allll the very familiar complex number system formulas etc that you have already seen that can be applied to other stuff you've already seen it will be intriguing and it all makes good sense. It's not rockets and stuff it's just capacitors and inductors and they can be explained down to every last electron flowing in and out if one wants to get that fancy about it. You have done all the math study work before and the maths operations are all known to you. You simply need to substitute the variables in the maths systems for electronic terminology and abbreviations and voila... With that math degree you have it'll be a blast looking at the stuff.

 

[catch22]

But, we are injecting only DC current (ideally), and C165 blocks DC. So it will have no effect on Q105.

I said "ideally", because we never have pure DC, there is always some noise, and we don't want to inject noise. That's why I sketched the second version, with additional RC filters from the +/-15V supplies, just in case they have any appreciable noise. They shouldn't, and almost certainly don't, but that's an extra precaution. Also resistors generate a small amount of electrical noise all by themselves, but the low value of R119 will greatly attenuate any noise that the additional resistor network would add.

It's the same thing. The small DC offset voltage we create is blocked by C107, C167, C101, C103, C165. The only paths remaining are through R119 to a local ground node, and through R109 to the base of Q103. So it really is very simple! A lot of electronic analysis is figuring out how to simplify the problem.

The flip side of the fear of the unknown is the excitement of trying and learning new things. I commend you for your adventurous spirit in considering modifying your Yamaha!

Thanks, missed from me at the first look that most of the "everywhere" will be blocked by the capacitors (in the ideal case of course)

The modern approach is to run thousands and thousands of simulations.

So computer modelling of any component, defining its behavior as function of time, temperature, voltage, current and so on, even including all real life limitations of materials, must be a solved problem? And then simulations run in order to model the behavior taking into account random variations (within tolerances)? Cool stuff!

Circuit design techniques have been developed over the years to reduce the sensitivity to these effects. For example, the base to emitter voltage Vbe of a bipolar transistor decreases linearly with temperature. This makes it hard to bias a single transistor so it conducts the same current at all temperatures. This is circumvented by using a pair of well matching transistors, such as the differential pair Q103 and Q105. Q105 buffers the negative feedback from the amp output at applies it to the emitter of Q103, as they have a common emitter connection. If an increase in temperature causes the Vbe of Q103 to decrease, it does the same to Q105, and the currents that each conduct remains unchanged. In effect the change in Vbe of Q103 is countered by a matching change in Vbe from Q105. This completely solves the problem of the temperature dependence of Vbe in this circuit.

Thanks, this is a good example for me to know that by looking at the circuitry closer, some bulding blocks serve the "main" task (for example the task of amplyfing the signal), while others are there to overcome specific constraints of the components itself

 

[catch22]

I will try and make this fast because I have barber appointment at 11 am in ~35 minutes. When you see a proper textbook detailing the operations of capacitive reactance and inductive reactance and see allll the very familiar complex number system formulas etc that you have already seen that can be applied to other stuff you've already seen it will be intriguing and it all makes good sense. It's not rockets and stuff it's just capacitors and inductors and they can be explained down to every last electron flowing in and out if one wants to get that fancy about it. You have done all the math study work before and the maths operations are all known to you. You simply need to substitute the variables in the maths systems for electronic terminology and abbreviations and voila... With that math degree you have it'll be a blast looking at the stuff.

Good stuff So, no magic Each component still can be modelled. Only that real life models are more complex comparing to theory. Because they are physical things and nothing is ideal in the physical world

 

[Doodski]

Good stuff So, no magic Each component still can be modelled. Only that real life models are more complex comparing to theory. Because they are physical things and nothing is ideal in the physical world

Back! No magic unless you believe in voodo and other assorted stuff...lol. Being as you are a smart guy with a proper education I trust you are not into that sort of alternative reality... Hehe... I have not modelled anything other than a plastic kit Revel Chevelle SS car model in my youth...LoL. I don't model electronics circuitry and I have no capability to do that in any manner shape or form whatsoever. What I did with electronics circuitry is what I chat to you about. Hands on direct contact stuff at component level analysis, trouble shooting, diagnosis and repair. Most 100% everything that I have seen was either broken or with faults and so the theory that I know and love is handy but the stuff was not operating properly and theory kinda does not work oftentimes and that's why it was brought to me. So... I had to combine theory with analysis of malfunctioning circuitry and fix it. It can be convoluted and tricky. One learns to deal with and use interesting and imaginative techniques to overcome and divide and conquer the repair. I specialized in mechatronics which is even more tricky than just electronic circuitry without a mechanism. Yes, nothing is ideal...LoL... You nailed it there for sure. Additionally after electronics repair became not economically viable I was able by being in very good physical condition and passing drug tests to transfer my mechatronics skills into electromechanical work manufacturing oil and gas heavy equipment like different types of huge earth drills, very specialized large wheeled and tracked vehicles, massive trucks that can drive through 6 feet of mud/muskeg all day with 40 ton loads on the back. A very physically demanding occupation. I was the wiring assembler guy as they are called on the floor where we built the stuff. Been a coordinator for a HVAC engineer while supervising the technical resources (me) for hundreds of HVAC tradesmen and gave them on demand technical support for electronic and electrical issues that they encountered with big and small HVAC stuff. Worked as the electronic QA/QC/service electronic technician for ~15 assorted engineers/mathematicians where we manufactured very high temperature rated downhole tools for oil and gas exploration, downhole earth strata studies and science oriented exploration. Many of the downhole tools where returned for analysis and service and seemed like they had been in a volcano as they where burned and stuff. Think like 177 Celsius rated/calibrated electronic circuitry remote sensing tools for up to a patented and calibrated 20K PSI pressure measurement and extremely huge torque and strain measurement tooling that where used around the world. So I never claim to be a software guy or a modeler of circuitry. I am a hands on guy that likes working with the peeps in a team effort and find office work to be mundane and boring. But I enjoy hanging at home doing stuff on the PC(s) like what I do here at ASR. That's a very abbreviated shortened rundown of wot I did. I wish I did not need to stop because it was getting more interesting for sure but I am retired now. My dream job appeared right before I retired. I had a telephone interview/conversation with a head hunter from Scotland that wanted me to go work with the huge trucks at the oil sands operations that where going to driverless operation. He like my resume because I could handle advising the various trades, manage the servicing of the trucks' technology and peripheral systems that enables the driverless operation and interface with engineers and other experts. I am certified for technical drafting, have a electronics study under my belt and many many hours as a component level electronic technician plus I am a registered first year millwright and first year instrument mechanic. Basically I applied for and challenged the trades exams, passed the challenge exams and was granted the first year accreditation for 2 different trades so if I ever wanted to do that trade I would get a head start. I pushed for 2 years accreditation for each trade but there where hang ups and I received the 1 year accreditations. It was something to do as a project so I did that for the experience of it. The head hunters and engineers like all of that combination of various stuff and I've had a bunch of offers for some very cool positions travelling and other stuff in fixed locations. I like to gab a lot....LoL. Hence I am here. I don't model stuff...LoL... Good thing this is a thread where I can be chatting or the MODs would reprimand me for blabbering...LoL...

 

[Zapper]

So computer modelling of any component, defining its behavior as function of time, temperature, voltage, current and so on, even including all real life limitations of materials, must be a solved problem?

Nothing is ever truly solved in engineering - someone is always pushing to do better. Same with modeling - there are always physical phenomenon that could be modeled better. For instance, the subtle details of capacitors and inductors that can cause small amounts of distortion that Amir's audio analyzer can detect are not properly modeled, at least in the publicly available models I've seen. Even binding posts can cause distortion. I recall one amplifier - one of the Buckeye Hypex amps if I recall, had a worse SINAD figure than expected in Amir's tests. It turned out that the binding posts were to blame - they had some ferromagnetic material in them (steel of some sort), which created a tiny amount of nonlinearity that was detectable in the measurements. I'm sure that wasn't modeled.

At work I design analog ICs, not PCBs. There the emphasis is achieving greater accuracy than ever before, so the models have to get better too.

 

[Zapper]

Many of the downhole tools where returned for analysis and service and seemed like they had been in a volcano as they where burned and stuff. Think like 177 Celsius rated/calibrated electronic circuitry remote sensing tools for up to a patented and calibrated 20K PSI pressure measurement and extremely huge torque and strain measurement tooling that where used around the world.

I would like to see that. I designed a communications chip for downhole, to pass data up and down to the cutting head. It had to have a 20 hour lifetime @ 175C iirc. No overcurrent or overtemp shutdown, it just works or dies trying. Had to package it in the old metal cans and ceramic DIPs because none of our plastic packages would take that heat. Each one sold for hundreds of dollars - high profit, low quantity.

 

[Doodski]

I would like to see that. I designed a communications chip for downhole, to pass data up and down to the cutting head. It had to have a 20 hour lifetime @ 175C iirc. No overcurrent or overtemp shutdown, it just works or dies trying. Had to package it in the old metal cans and ceramic DIPs because none of our plastic packages would take that heat. Each one sold for hundreds of dollars - high profit, low quantity.

That's kinda like the situation I experienced too where we sold thousands of expensive small diameter downhole tools at a time to corporations and even sheity little countries run by dictators that had petroleum resources. We would dynamically qualify (ICs where electrically powered on via lotsa wiring leading into the oven.) in a temp calibrated oven many at a time the ICs used in the downhole tools and those that survived went on through the process to become calibrated downhole temp and pressure measurement tools. The pressure sensors came from a Scandinavian country. The special O-rings where all very high temp rated and regulated by the government and we could not export to specific countries for fear they would apparently use them in nuclear research or related stuff. The tools where encased in a hastelloy stainless steel in-house machined tube with standard drilling threads for mounting the tool to the drill stem. The tools where if memory served me correct ~2" in diameter (I had a German made high accuracy metal lathe at my work station and I lathe canned the old tools pretty often to refurbish them so the diameter was not absolutely critical but the thickness of the can sides was critical.) and up to a 8" hexagonal profile for the bigger really rare and very expensive downhole tools. Those where made in limited production quantities. I exclusively assembled them and reported directly to the head engineer about them as they where his pet project and he held the patents for the technology used in the torque and strain measurements performed by the tool. They also measured temp and pressure. There was a lot of stuff that I had to very carefully and very neatly mount inside the big heavy tools and it was critical that I wired it immaculately because the design relied on the flex of a mylar sheet with copper coils in the mylar that was adhered to the downhole tool sides internally and as they flexed from torque and strain the reactance change was used for the measurements. I cut my hands so many times on those tools because the drill stem threads where extremely sharp and handling the 8" tool required brute strength as it was fairly long.

 

[catch22]

Back!

Haircut done? I always feel better after, cannot really explain why!

Additionally after electronics repair became not economically viable I was able by being in very good physical condition and passing drug tests to transfer my mechatronics skills into electromechanical work manufacturing oil and gas heavy equipment like different types of huge earth drills, very specialized large wheeled and tracked vehicles, massive trucks that can drive through 6 feet of mud/muskeg all day with 40 ton loads on the back. A very physically demanding occupation. I was the wiring assembler guy as they are called on the floor where we built the stuff. Been a coordinator for a HVAC engineer while supervising the technical resources (me) for hundreds of HVAC tradesmen and gave them on demand technical support for electronic and electrical issues that they encountered with big and small HVAC stuff. Worked as the electronic QA/QC/service electronic technician for ~15 assorted engineers/mathematicians where we manufactured very high temperature rated downhole tools for oil and gas exploration, downhole earth strata studies and science oriented exploration.

This is very exciting to see you've got some unique experience, and also that it was demanding yet for sure fulfilling! It covered both using your brain and physics and this is great. I do IT for living, it always was natural for me, I understand logic well, either because of the math/programming background, or rather because of logical mindset that brought me to math/IT in the first place. Yet I always was founding it limiting because of absense of physical aspect. (I try to compensate lack of physical involvement at my job by DIY, mostly woodworking. No decorative stuff, but utility oriented like bespoke cabinets, and for example 2 large sheds in my backyard (one is 4m height) I built start to finish all by myself).

There was a video in this thread posted earlier, an old video around 1960x, about soldering technics trained in NASA for stuff they sent into space. It was interesting to see how high they pushed the standards. (And also you see how good was video content before affilliate links vere invented hahaha I reckon in your job you also dealt with stuff with very high requirements for reliability.

 

[catch22]

Good thing this is a thread where I can be chatting or the MODs would reprimand me for blabbering...LoL...

Hahaha we have an excuse, we are waiting for delivery of the caps to start the repair

 

[Doodski]

Hahaha we have an excuse, we are waiting for delivery of the caps to start the repair

Thankssss.

 

[catch22]

Nothing is ever truly solved in engineering - someone is always pushing to do better. Same with modeling - there are always physical phenomenon that could be modeled better. For instance, the subtle details of capacitors and inductors that can cause small amounts of distortion that Amir's audio analyzer can detect are not properly modeled, at least in the publicly available models I've seen. Even binding posts can cause distortion. I recall one amplifier - one of the Buckeye Hypex amps if I recall, had a worse SINAD figure than expected in Amir's tests. It turned out that the binding posts were to blame - they had some ferromagnetic material in them (steel of some sort), which created a tiny amount of nonlinearity that was detectable in the measurements. I'm sure that wasn't modeled.

At work I design analog ICs, not PCBs. There the emphasis is achieving greater accuracy than ever before, so the models have to get better too.

Those examples of Current Mirror and the other one (for overcoming termal dependency) show how analog electronics is different from pure logic.
The circuitry contains not only the flows, serving the main purpose of the application, such as for example "amplification of a signal", but also in itself contains "clusters" of components that do some tricks to overcome its own limitations. This is not needed in pure digital domain. And this makes analog domain more interesting.

Although generally speaking, what makes any engineering interesting, is that it aims not just for a solution, but for a solution that is created within constraints. Size, weight, cost of materials, cost of production, etc.
Constraints are driving the progress.

 

[Doodski]

Haircut done? I always feel better after, cannot really explain why!

Haircut is completed. High and tight is the order of the day.

This is very exciting to see you've got some unique experience, and also that it was demanding yet for sure fulfilling!

I never ever planned it. It was all just the progression that naturally occurred due to various opportunities available as I move from job to job.

Yet I always was founding it limiting because of absense of physical aspect.

Before I took up electronics studies I was a salesperson and did very well but I was a skinny, tall guy with little muscle. So after I got things going with the electronics work I had extra hours and I fell in love with in-line skating. I skated around and around a very large park on asphalt paths ~50k km in the afternoons over a period of years, worked out in the gym building muscle 3-5 days a week and swam in the pool after each weight session and rode the stationary bike for 45 minutes for cardio before weight lifting. I cross trained that way for years and never chose to get really big I did chose to build my core to be very solid. So I was body weight 174 pounds and benching 177 lbs and maxing most all the machines at the gym except the strong machine. Very defined abs, back like a boxer, square chest, arms like a basketball star, legs like a track athlete and so forth. It was a mish mash that worked well for me and is still there inside the fat and just needs to be toned again if I can handle the pain of that...LoL. I'm 300 lbs now and 6 feet tall. If you are going to work out you need to make it fun and interesting and develop skills by movement and sport. Simply pushing weights is boring and does not develop skills.

(I try to compensate lack of physical involvement at my job by DIY, mostly woodworking. No decorative stuff, but utility oriented like bespoke cabinets, and for example 2 large sheds in my backyard (one is 4m height) I built start to finish all by myself).

Excellent stuff. It achieves things for you and gives one a sense of pride, accomplishment and worthiness. If I where you I would continue with that and build ever more.

There was a video in this thread posted earlier, an old video around 1960x, about soldering technics trained in NASA for stuff they sent into space. It was interesting to see how high they pushed the standards. (And also you see how good was video content before affilliate links vere invented hahaha I reckon in your job you also dealt with stuff with very high requirements for reliability.

I attended a full time and expensive PACE Soldering study in Canada and became certified. It took me a month of my time for the course that was held in a very large mobile trailer designed for this and full of lotsa very cool soldering equipment of all types and prices. It was government radar systems specialists, engineers and me the only technician there. I laid waste to that course as it was pretty easy and I shocked and startled the peeps there when they saw my soldering skills compared to theirs...LoL. Basically I was a bench tech and soldered all day everyday and they often replaced PCBs and did not need to solder very much and so I had a leg up on them that way.

US Government & Military Support | Pace Worldwide

PACE Provides Equipment, Technology & Training to All Branches of US Military & DoD Since the 1960's, PACE Worldwide (also known as PACE Incorporated) has had a long and rich history supporting the US Military and Department of Defense.

paceworldwide.com

Certified PACENTER® Training Centers | Pace Worldwide

paceworldwide.com

 

[catch22]

Executive Summary:
Today the replacement cap arrived, I desoldered the old one and..........

I'm proud to confirm that I have SUCCESSFULLY done my first ever electronics repair!

(Well, not the first in general, but definitely the first ever "Smart" repair, where the cause was not obvious to the naked eye and that could only be done after analysis from knowledgeable people!)




Longer Version:
I've got 5pcs of new Nichicon 0.022uf, out of curiocity measured them all, they were 19.3, 19.5, 19.5, 19.6, 20.0nF which is at room temperature 10% lower than nominal.
I know they will increase capacity at highter temp, so when warmer they will be much closer to nominal maybe within 5% or less. So, de facto they are much more precise than allowed tolerance of 20%.

I cleaned the back of the board with IPA, desoldered and measured the old guy, and it was just 4.5nF, less than a quarter of nominal - "absent" as someone here said.
I didn't remove the board but have checked the visible are for possible cracks, all didn't see anything else suspicious.

I soldered the new cap, and connected the power and voila!

It knew that L/R imbalance of the DC offset would not change (because it is likely caused by the issues in one specific channel, and not by the power that is shared for both).
I checked it anyways, on the cold amp it is R:17mA / L:73mA (and slowly creeping up), so, essencially will be the same as before.



@Doodski @Zapper @Dimitri @Glint your opinions were absolutely right about the main suspect C254.
Tagging you FYI, of course you knew that from the beginning
Thanks to you and to all ASR people for your support!

@Doodski @Zapper Give me a couple of days to enjoy and harvest the results of this success before I break the amp forever in the next risky vernure of modding it for bias adjustment

 

[Doodski]

Executive Summary:
Today the replacement cap arrived, I desoldered the old one and..........

I'm proud to confirm that I have SUCCESSFULLY done my first ever electronics repair!

(Well, not the first in general, but definitely the first ever "Smart" repair, where the cause was not obvious to the naked eye and that could only be done after analysis from knowledgeable people!)

That my friend is a great experience and you knew you could do it and just needed to get 'er done!

It knew that L/R imbalance of the DC offset would not change (because it is likely caused by the issues in one specific channel, and not by the power that is shared for both).
I checked it anyways, on the cold amp it is R:17mA / L:73mA (and slowly creeping up), so, essencially will be the same as before.

Yeah, that is going to be a fun thing to track down for sure. Not as easy as the repair you completed in the standby power supply section.

@Doodski @Zapper @Dimitri @Glint your opinions were absolutely right about the main suspect C254.
Tagging you FYI, of course you knew that from the beginning
Thanks to you and to all ASR people for your support!

I for one enjoy this stuff and get a kick out of spreading good cheer and assisting peeps with their electronic endeavors. So thanks for letting me be one of the Doods that assisted.

@Doodski @Zapper Give me a couple of days to enjoy and harvest the results of this success before I break the amp forever in the next risky vernure of modding it for bias adjustment

Yes, that is also the best route to proceed in because to test it thoroughly before any other service work is the smartest thing going. Well done!

 

[Glint]

hurray! big congrats on doing your first repair!