Did my AVT-8710 die?
3 Attachment(s)
I looked at the user's manual, but it doesn't say anything about this combination of lights even being a possibility. All lit except "Color". I kept unplugging and re-plugging the DC barrel jack, holding Reset, attempting to clear the error. Eventually the lights went back to normal, but the video output was just grey instead of proper video. I can't even get it to display colorbars, which is what it's supposed to show when there's no video input.
Attachment 15025 Capture with vertical delay added, to see VBI. Just grey... Attachment 15027 Green model, purchased second-hand via Craigslist in late 2013. |
Open it and look for any signs of smoke, capacitor leaking or bulgy, check the voltages with a voltmeter. While in operation touch the chips and see if any are overheating, Get a can of compressed air and spray on the chips, if you get picture that means they are going bad.
|
Thanks for the suggestions, latreche.
It's working again, after many power cycles. I have no idea what I did differently on this most recent attempt that changed things. This time, I took cell phone videos as I was disconnecting and reconnecting the DC barrel plug, because I wanted to prove to myself that I wasn't going crazy and that this thing really is exhibiting random light patterns and other odd behaviours. I'm especially enjoying listening and re-listening to my "WTF?" reaction when it suddenly started working for no reason. |
Alright, first off, don't worry yet.
Yet. I've seen at least 50 green AVT-8710s in the past few years, and probably closer to 100 in the past 20 years. What you're seeing is a natural breakdown of the device, natural progression. I've seen everything from flawless units, to early issues, to late issues, to failed units. There's really no caps that affects anything, and few caps whatsoever. The problem here is the chips. Over time, those overheat, and fry themselves. This model TBC was nice and cheap (below $500, when most TBCs were at/over), but had cheap components, namely the small plastic case with no real ventilation. Combine that with no heatsinks, and problems. It was mostly user error, pushing the unit past 6 hours of continual use. Often continuously, sometimes even leaving the unit on for days (which I did by accident once). Depending on the problem chips, it can be replaced. Others, like the FPGA, are bitlocked. So when it goes, unit is dud. Thankfully, it's usually not the FPGA, but the simpler ADCs/DACs. Your unit now has what I refer to as the AVT-8710 "attitude problem". Do not quickly plug/unplug, even attempt it without a few minutes break in between. In some cases, the unit seems to want to be "primed". So plug (see errors) and unplug is one prime. Sometimes it can take several attempts ("primes") to get it cooperating, with rest between each attempt/prime. If priming doesn't work after 5-6 tries, with several minutes between each, then let it rest for at least 30 minutes before reattempt. Do not spray "air" on the chips. That "air" is a liquid with propellant that can do damage. Linus (LTT) would have a fit if he saw that your unit still had the peel. :laugh: FYI, Goo Gone will clean off that sticker residue, but then use alcohol to remove the Goo Gone residue. DO NOT get any liquid in the case, so ideally dismantle it all if not extremely careful, using Q-tips. Over time, this "attitude problem" will get worse. Again, progression. For most casual users, not an issue. But I assume you're more like me, and continually use TBCs on a regular basis. It will start to piss you off after a while. I'm really busy lately, but wanted to take some time to answer this for you. You're fine, don't worry. Just proceed carefully. And realize it may get irritating with regular use, aka time to add another TBC. |
Wow.
Are these the units where there's uncertainty about the pack-in PSU? Some have about +3VDC on others? If it's using a linear regulator that's some serious heat difference, potentially Watts of heat difference. I can't find a clear picture to determine what's going on, but there's at least 85 capacitors inside the unit looking at the picture I found, so again, we can't really say this unit has 'few' caps. |
Quote:
15V, 600mA only 12V, 500mA works (yikes?), 800mA to 1.0A works (low?), 1.2A to 2A usual PSUs used. Be very careful with polarity reversal cables. Some do not work, accelerate "attitude problems", 50/50 odds for good vs. bad. It's always been my understanding that the 600mA rating of the 15V kept it from drawing too much power, But the 12V needed more amps, and had no limit. Quote:
Or "picture", as in you need a better photo? There are some on the forum, but I can take a high res clear photo of multiple AVT-8710 units. What exactly are you needing to see? Quote:
Yes, there's at least 50 small caps, but those appear to all be ceramic. It's extremely rare for a ceramic cap to blow. (And then a PITA to replace.) The AVT-8710 was the lowest cost (I don't want to use the word "cheap" here) TBC made, that was still quite powerful and effective. At least during the "green" era (not "black"). But it doesn't mean they used cheap parts inside. In fact, it was the more expensive DataVideo that used cheap electrolyte caps that have failed en masse starting in 2019. |
Do you mean electrolytic?
|
Quote:
electrolytic = relating to electrolysis or an electrolyte So ... yes? :unsure: There's only 6 aluminum electrolyte caps. |
Quote:
There are six round SMD packages on the board from what I can see, and you're right to identify these are aluminium electrolytic capacitors - however, these are far from the only capacitors that fail. You're right to say that ceramics have a low failure rate in service, although it's well known that they do 'crack' with vibration or thermal shock, that's unlikely in this scenario though. Tantalums - that's another story, which also come in packages, which in this application look like 0402 to me; but I'm only looking at a blurry photograph with no scale, and only one of my eyes works properly. Quote:
Quote:
What do you mean by 'had no limit'? That's the bit that's thrown me, and are we speaking of current or power? They're entirely different concepts. I might be able to understand if you explain a bit more and we can help you with the terminology? Remember, and I think this is where you may be getting a tad confused, this device is highly unlikely to operate at either 12V or 15V internally, it will operate at 5VDC, 3.3VDC or a combination thereof. The 12/15V is the power supply, and the device will regulate this internally to whichever Voltage is required. Depending on the way the device regulates power there is a potential for excessive Voltage to be simply consumed as heat, essentially the voltage regulator has to turn into a heater to dissipate Voltage - it's not very efficient. The difference in dissipated heat values between 12V and 15V is quite a lot. This is what is meant by 'linear regulation', it's cheap, old fashioned, bad for the planet (and equipment) and does require that the input Voltage is to spec', it's not a case of 'more Voltage-more better' to coin a colloquialism that nearly fits, but 'more Voltage - more heat'. They maintain nearly ripple-free current, however, and they're very simple to implement (three components). They're also just about noise-free (for the purposes of this discussion.) It could be a DC-DC SMPS, which mitigates the sort comings of linear supplies, but I can't determine that from a blurry photograph coupled with not-fantastic eye-sight, but judging by the inductor this would well be the case - but then we're back to SMPSs which are fairly Voltage agnostic and we go full circle back to power. There's no such thing as a free lunch, we're getting noise instead. If you've got a board, have a look and see how the power regulation works, it may be giving you a huge clue as to what this failure is... If you're not sure, if you post up a clear image of the board, we can probably take some educated guesses as to what is going on. |
Quote:
If you want to be an ass, and play semantics, I have more important things I can be doing. And I'll just go do those, and this conversation can end. I'm in no mood for trollish Twitter/Facebook style internet bullshit. I see this sort of behavior for what it is: an attempt to feel/seem superior. Now then, the AVT-8710 has some mysteries. One of those is the PSU, 12V vs. 15V. If you want to have a serious discussion about that, free of snarky comments, then we can proceed. Pick one. My next post here (or non-post) awaits your response. :hmm: |
I don't know what to say really, if you want to use strange terms go ahead, I'm sure you'd be more than happy to admonish somebody for using incorrect terminology. That's not trolling by my estimation.
I'm trying to have a legitimate discussion, I'm proposing we take a look at the board and try and work out what may have gone wrong and what might be repaired? I have even offered to try and help you understand what the issue may be. You have spent decades on the internet correcting people in your area of expertise, and as you've also said in your defence, 'They need to be educated, not coddled.' - I'm trying to stop you making a bit of a silly, albeit inconsequential error. It is a strange term to use, that's all? Sorry if you feel so insulted, but I would have thought your character, after years of being 'rather blunt' with strangers on the internet, would just accept that. |
Quote:
Technically "caps" is also incorrect, no? :laugh: ^ /end this OT sub-topic :warning: Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Do you think there is a correlation between the "attitude problem" (degradation of the unit, due to heat) and the 12V/15V? I'd never consider that. Hmm. I doubt it. But still, hmm. :hmm: Sadly, this conversation is too late for most of the units -- long ago fizzled out, damaged beyond use. But for those of us still holding and using, good info to keep it going. There's always potential to customize the unit, add heatsinks, even replace some parts, use better PSUs. Anything to make it last another decade or two! |
Quote:
Your concept of electrical current is still very muddled, you've broken fundamental laws of the universe with your explanation. Current is drawn, and we can model as a Thevenin equivalent circuit, this isn't up for debate really, although you'd probably win a Nobel Prize in physics if you could prove what you describe, in which case, I'll 'take the L' on that one. Imagine this, your car has a battery that can say deliver around 650A instantaneously at 12V for starting, the same car with have a oxygen cell in the emissions control circuit that may draw say 5pA - they can both be connected to the car battery, a device will only ever 'draw' current, as current is what flows when a unit of potential flows through a unit of resistance. And yes, we can determine a tremendous amount by looking at the PCB here, we can probably work out what has failed as this sounds like an absolutely classic electrolytic failure in an SMPS and is probably very easily solved for a few pence. We can also determine if using the 'wrong' PSU is leading to excessive heating or whether this isn't a risk. This is I think is connected to the 'priming' you describe, I would imagine you are forcing capacitors to reform and allow the SMPS to operate at the correct Voltage as a first pass guess, until then it's probably grossly under/over voltage. When we combine this with knowledge of the units generating excessive heat in a poorly designed case, electrolytic failure is a great place to start - and if using a 15V PSU is also causing additional heat, then it's a very logical line of enquiry. But until we can see a clear image of the board, it's all just guesswork. |
Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.