Everybody, I feel your pain. Perhaps I can shed some light but it is up to you to decide how you want to handle your Onkyo problem.
Be aware there are at least two distinct problems being detected in these replies, IMO.
1) HDMI capacitor thermal failure
2) HDMI cold solder failure
The following statements are ONLY my opinions and observations! Please bear in mind, being human (or a close facsimile) I am prone to human error.
I have observed most or maybe all of these failure symptoms personally on multiple units and repaired one of them myself so I am personally confident in my relative accuracy. For the cold solder failure I might have not (yet) experienced, I am passing on here what a quick search of AVS forum teased out.
The first, and most vexing, problem is the (inevitable IMO) heat-induced HDMI capacitor failure that shows up with increasingly frequent intetrmittent failures on the HDMI, encompassing both audio and video problems of all sorts that eventually become permanent, such as:
A/D video upscaler (component/composite input to HDMI output) fails
On-screen display/setup menu fails to display through HDMI
Copy-protected playback (e.g. bluray) freezes
HDMI audio dead and video output corrupted/random noise/no sync (dead)
HDMI remote device controls (power on, input switching) unpredictable failures
Onkyo has never officially acknowledged this as a design defect to my knowledge although IMO it is clearly a thermal design defect that might cost consumers millions in aggregate. In some cases of very early failure some customers have reported that Onkyo will repair or replace the HDMI board for free if the warranty recently expired, the customer is sufficiently saddened, and the service representative is sufficiently empathetic, or when the moon is in the seventh house and Jupiter aligns with Mars?
These failures come and go seemingly at random and, at first, just switching the input selector away and back to the problem input can temporarily fix it, as can calling up and dismissing the on-screen setup menu, re-plugging or plugging the component into an alternate HDMI input or using an alternate component on the same input or 'resetting' the unit. The temperature in the ambient environment can also affect the problem. Eventually the problem becomes permanent and affects the entire HDMI function.
All of these problems usually stem from noisy power supply on the HDMI board due to the low-temperature rated (commercial-grade as opposed to mil spec) surface-mount electrolytic power supply filter capacitors being cooked with too much heat dissipation from ICs on the HDMI board and no heat sinking or fans to cool the chips/board.
Any repairs with original equipment (Onkyo, are you listening?) will almost certainly fail again over time unless modifications are made to either the way the unit is used (inside a refrigerator? in a very cold climate? in front of a desk fan?) or the way it is constructed.
These HDMI failures are well-documented here on AVS Forum and on Youtube and ebay and the Onkyo technical bulletin from the TX-SR606 HDMI capacitor failure has been leaked to the Internet. DIY repairs/modifications noted here at AVS include replacing the failed capacitors, changing the capacitors with higher-temperature rated and/or leaded capacitors that stand off the board (to improve the grade of capacitor and/or reduce thermal conductivity from the board to the capacitor), and installing custom heat sinking on the ICs and/or constantly running cooling fans inside/on top of the case to keep the board (and its capacitors) cooler.
I removed five of the capacitors on my TX-SR606 and replaced them with forty-year-old 'surplus store' leaded capacitors with double the voltage rating and approximately ten times the physical size, that had been kicking around in my junk box since I was a kid. I used the instructions I found on Youtube and a personal web page that had a copy of the leaked Onkyo technical repair bulletin for the TX-SR606.
Note that in order to perform the repair safely, before removing the HDMI board two pins on the board must be soldered together to prevent static discharge from blowing the board. The pins are clearly marked on the board of at least some units. I did not know this at the time and may have permanently fried the upscaler and on-screen menu function.
Note that in some cases, the upscaler and on-screen menu will not recover after replacing capacitors on the HDMI board, indicating that capacitor failures on the analog video board may also be a factor, or perhaps the ESD vulnerable pins were not soldered before the board was removed.
These HDMI power supply filter capacitor failures cause loss of signal integrity from noisy local regulated power supplies on the HDMI when the electrolytic filter capacitors become so degraded they no longer filter power supply ripple effectively. The failures start out intermittently when random bit errors or bursts cause data corruption. For example, if the copy protection encryption key handshake gets a bit error and the key returned by your television becomes corrupted/invalid, blu-ray playback instantly freezes until the connection is reset, then it freezes again some time later.
The second failure, I have no personal experience with as far as I know, although I might already be having it intermittently on my TX-NR929 (or maybe it is something else causing intermittent loss of audio such as bit errors in the audio packets from the capacitor failure). I did read up on it here on AVS Forum though so I have some information to pass on.
It is apparently related to cold solder on the BGA balls of a DSP chip on the HDMI board and may appear suddenly as a hard failure due to thermal expansion/contraction breaking the contact of the balls from the PCB.
This failure is also intermittent in some cases and permanent in others, probably depending on if the broken balls are making intermittent contact or if a subset of parallel-connected power supply/ground balls has lost contact and compromised power supply integrity inside the chip. Ambient heat could be a factor, as can time duration since last powered up. External component factors such as connected devices or network are unlikely to affect this failure. It appears that the affected DSP handles both audio and network functions and is euphemistically referred to by Onkyo as a 'defective network chip' ostensibly blamed on a supplier as opposed to a manufacturing defect by Onkyo.
They probably had a batch of DSP chips with defective solder balls, a flux quality/application problem, or incorrect reflow setting on the oven, either temperature or duration setting.
My guess is, Onkyo management are becoming sensitive to the HDMI capacitor thermal failure mode and desperately seeking a 'quick fix' to reduce its frequency since it is such a black eye. Probably, they experimented with different fluxing etc. and/or reduced the oven temperature or duration (or both) to try to reduce the initial thermal shock to the affected capacitors, and created a cold solder situation in the process.
I have contacted Onkyo to request they repair their (IMO) defective HDMI thermal design and I presume others have also. Probably they have people reading these AVS Forum postings and seeing my critical comments along with others. I have been more pointed and vocal and analytical, a real PITA, so maybe this cold solder situation is my fault despite my lack of authority and influence and prestige? Or maybe I give meself too much credit for my influence.
This speculation of mine is of course purely a guess, but if I were an amoral CEO looking for a low/no-cost quick fix with no short-term impact on the quarterly financial results or stock price, this sort of mucking around with the soldering is the type of boneheaded solution I would squeeze out of my quivering manufacturing engineers, rather than redesigning a congested backplane to make room for more cooling or using higher grade of capacitors. Reducing the applied thermal energy would definitely affect the reliability of larger components (such as the DSP chips) more because they have more thermal mass and take longer to heat up. This would also be more problematic with the lead-free ISO requirement since lead-free solder is more difficult to reflow but that is old news now that everyone has gone lead-free long ago.
I have expressed my opinion on AVS Forum that the decision to drop Audyssey is also probably related to the congested thermals on the backplane and the need for additional DSP power for Atmos object-oriented sound format. Something would have to go if there is only room for so much dissipation back there.
Perhaps the balls on the supplied chips were defective, or maybe all my second-hand information is totally wrong. I may be full of puckey on this issue.
I have personally participated in such maneuvers on products I helped design, including one high-speed multi-lane serial interface we farmed out to a contractor that failed with random bit errors due to pushing the speed above the silicon process spec and having a slightly too-heavily loaded clock net. The 'solution' was to crank up the local power supply voltage to the absolute top operating limit, put additional capacitors around the IC, and develop worst-case test vectors that screened out parts likely to fail (approximately a 40% yield hit).
Mucking around with it ended up taking longer to fix and more expensive redesigns than just reducing the loading on the affected clock net with a simple relayout, IMO. Then again, I did not look over the affected block personally so maybe it was just too difficult to fix, especially since the contractor was already released. Maybe we were designed into a corner, just like Onkyo may be designed into a corner.
I protested the decision but I developed new test vectors anyway. I eventually had to write a vector that ran repetitive partial chip resets at-speed, something that never happens in actual use but that loads on-chip power supply dramatically and teased out the failure mode from marginal parts by dousing the on-chip power supply and corrupting the clock signal, thus glitching the parallel-to-serial shift register, and desynchronizing the data stream, making the parallel bits either flip state or develop a permanent shift to the wrong serial lanes, a malfunction that persisted until the serial interface was reset.
I was also the harbinger of bad news to the engineering lab. Apparently management was playing this one close to the chest but nobody bothered to tell me so of course when they finally put me (back) on the case I blabbed to the hapless validation engineer who was sitting in the lab frustrated. I took his ire gracefully when he learned the truth. He had been staring at corrupted data for months and playing around with bypass capacitors trying to diagnose it in isolation.
I had already diagnosed this failure in digital simulation before chip tapeout months prior but management told me it was a 'simulation artifact' according to our PhD. The IC test engineers had turned off the monitoring of that particular interface when they saw garbage out. We never detected the failure in the lab until that more advanced mode of operation was finally being checked in prototype circuit boards. When another PhD finally confirmed my digital simulation diagnosis in analog simulation I was vindicated but no one thanked me for my efforts to prevent the fiasco or my developing a crucial part of the workaround.
Note the company was already in the process of being sold at the time so management really did not care what we did to band-aid it as long as we did not affect the schedule. This was one of two known design defects in the chip introduced by contractors (the other was a bandgap voltage reference thermal failure from a nonsensically connected 'spurious' transistor), both issues originally detected and diagnosed by me, both worked around rather than repaired with a simple redesign.
These sort of design issues occur all the time and the challenge is detecting them and fixing them before they develop too much inertia.
When I see Onkyo apparently engaging in these sorts of band-aid approaches and not even being as effective as my former employer was in solving the problems, I have to wonder what the long-term vision of the company is (if it even has one). Lately it seems that corporations get abandoned by their founders and taken over and eaten by predators rather than handed off to long-term caregivers. Then again I am a jaded old bird so maybe I am just overly cynical.
Please note that the Onkyo web intake form for the 'defective network chip' problem seems to include questions referring to symptoms of operator error, external component failures, HDMI capacitor failure, and cold solder failure. For the purposes of this 'non-recall recall' (one-time warranty extension to repair cold solder?), your answers to the questions may affect whether you are offered a free repair. Presumably if you have both symptoms of HDMI capacitor failure and also cold solder failure, they will repair or replace your HDMI board for free, but if you also have clear symptoms of operator error checked off (or no symptoms of cold solder) they might assume you are just neurotic or experiencing a 'non-covered' capacitor failure.
How they handle intermittent failures they cannot reproduce depends on the technician doing the servicing, if he/she applies heat and cold spray and mechanical stress (bending/twisting the PCB) or on the phase of the moon.
Intermittent failures can be difficult to reproduce. Some technicians may assume failure where none is evident and change out the board, and some may just ship the receiver back NTF (no trouble found). Some might diagnose and repair only capacitor failure even though there is undetected cold solder, or vice-versa. I do not know if Onkyo unconditionally authorizes free repair in advance or retroactively adjusts the charges if no 'defective network chip' failure is found. Depends. You pays your money and takes your chances.
My recommendation is to prepare your short and simple speech in advance and call Onkyo rather than using the web intake form.
You might get better service if you fit their 'parameters', especially if you stress that the failure is highly intermittent and might not be reproducible. Maybe then the tech will have mercy on you, if the customer service rep actually forwards your comments. It might be better to hand-deliver your unit to a local repair center and bring a printout of my comment, although if your tech has already repaired a couple of these he/she is likely familiar with the relevant issues already (although probably disinclined to discuss it openly with you).
OK now that I analyzed it to death for you, good luck. Some enterprising individuals have offered unauthorized repairs on ebay at cut-rate prices, but I would steer clear unless you get no satisfaction from Onkyo and cannot competently DIY the repair in person.
I recommend that everyone who is personally affected by Onkyo HDMI issues send a registered letter to 'CEO of Onkyo corporation' expressing your disappointment with the quality of their products, particularly the alleged thermal management design defect of the receiver backplane.
Last edited by CherylJosie; Today at 12:56 PM.