This article is about one of my recent spontaneous projects. A few days ago, I got lucky on an ebay auction and picked up a broken Acer K330 projector. I often look for these kinds of offers because I have spent several years in the audio/video repair business and am pretty confident in my skills when it comes to fault-finding. So I figured, saving some money by restoring a broken device couldn’t hurt.
First some facts and features: The K330 uses a three color LED module, which promises a long lifetime and low energy consumption. A Texas Instruments DLP module handles image generation. In sum this lets me hope for good contrast and strong colors, even if the brightness of 500 lumens doesn’t seem that much. An interesting thing about this DLP chip is that it uses an uncommon diamond pixel grid for size reasons. Diamond in this context means that the pixels do not form a rectangular pattern like in the usual TFT monitors but rather a grid of 45° rotated and slightly squeezed, not completely rectangular tiles. Of course, this means that internal resampling has to occur to map the image from the rectangular domain onto the diagonal domain of the IC.
So, this projector was obviously broken when it arrived – what a surprise. The seller had already informed me that it had suffered from overvoltage of unknown cause. The VGA picture was supposed to be very green-ish, the HDMI input dead in whole and the media player erratic. A slight flickering in the picture was also mentioned. I’ll take you through the repair process for this device from here on.
Arrival / Testing
I first plugged in the device in and turned it on, since it wasn’t supposed to be totally dead. This is usually not a good idea if the device has a known power supply fault since things might get worse when the high voltage hits damaged parts. I decided to trust the seller on this part, and as expected, the projector started up in a few seconds and decorated my wall with a bright and colorful standby screen. No obvious dead pixels and all three main colours were present, which meant that the digital mirror device (DMD) and the LED light source including the drive circuits for both probably survived. So far, so good. The power supply could still be damaged of course, but I didn’t expect that. From my experience, overvoltage failure on switchmode supplies (SMPS) is pretty much binary failure. The good ones have a varistor protection following the input fuse, which means the fuse blows when the varistor trips – sometimes together with the varistor. No big damage, but it’s dead. The cheaper ones do not have the varistor and sometimes a bad/slow fuse, which prevents shutdown and causes the primary rectifier to fail, the primary smoothing cap to blow up, or kills the drive circuit. Also, dead. This one was running smoothly, so it should be fine.
Then I also tested the inputs, but nothing new there. VGA only green, HDMI nonresponsive, media player behaves erratic, hangs and crashes. Composite works, though, but with the horrible image quality you’d expect from such a technological dinosaur. Also present is the slight flicker, but only when selecting HDMI in – strange!
Let’s crack this thing open.
The usual word of warning: Whether you do or don’t try the following is your responsibility alone! Usually, repairs of such devices are not for the hands of unexperienced people, which is not meant to be a discouragement in general. Just be aware of your limit and remember that live current can and will bite you if you are not careful and certain in your procedures.
While examining the case, I noticed that someone had already tried opening it before. Unfortunately, he seems to have made a bad choice about his point of entry and lifted up the white top section of the case, which is supposed to be welded to the rest. Well, the few scratch marks can be fixed and the cosmetic damage is not too bad. This is how it’s done the right way:
- Turn over the device. Remove the two lightly glued rear rubber feet by prying them out at the edge with your fingernail or a plastic tool. Also remove the two circular silver patches on the silver border toward the front of the device. Remove the four screws you find beneath.
- Pull the bottom plate toward the rear side of the projector. It should slide a few millimetres, then you can pull it off.
- Use a small screwdriver and pry up the top panel of the projector along the upper side of the first slit in the silver outer shell! NOT (!) the white part at the top, it is welded on. Force the silver top up with your fingers at the front left/right edges, then slide your tool into the slit near the holding tabs as a lever and push the silver side panel outwards to release the tabs. See the picture below to see how the locking mechanism works. There are five tabs inside, one at the front center, two at the left/right centers and two more aligned towards the rear of the case. Once all are unlocked, leave the lid in place.
- Pry out the holding tabs at the rear side carefully using only your fingers! Once all are removed, pull up and back on the top panel until it snaps off. NOTE THE FRAGILE RIBBON CABLE!
- Remove the ribbon cable. Note that the projector works without the top panel attached, using only the remote.
This is what it looks like on the inside:
The big blue board contains all the digital logic while the small one contains the power circuits for the lamp unit. To remove the logic board, first unscrew the LED power board and carefully pull it up. Note the small pin header connector (LED control) and the locked power connector at the bottom. Unplug these two, leave the LEDs connected and push the board out of the way. Now unscrew the four visible screws on the logic PCB and the two bolts securing the VGA port to the case backside. Stick your finger in the small space visible between the speaker and the fan (exactly centered on the right side of the case in the above picture) and pull the board upwards gently. The two PCB mounting screws behind the lens assembly mark the spot where the DMD connector is at the bottom side (see following pictures), it should unplug easily. Take care when remounting! Finally, carefully unplug the power connector (top left, see next picture). Depending on your plans, you might need to remove the remaining connectors, but don’t worry – the sockets on the PCB are marked, and it is pretty obvious which goes where.
Hint: When remounting the logic board, take care of the calibration/check light sensor! It is mounted on a small PCB below the logic board, right on the light channel. It has a three wire cable which is pretty fragile. If the device turns on, beeps several times, turns off and signals “LAMP” failure, check the cable where it is soldered to the PCB. It can easily short out and will produce the described symptoms.
The supply is underneath the whole logic unit, and, according to a visual check, no overheating or capacitor-splattering has occured here.
Let’s start off with a brief function block overview of the logic PCB since no schematic is available so far. I mostly “read” that from the circuit by looking for datasheets of individual ICs, then sorting out which interface goes where and more along those lines.
Fixing the VGA port
First up is the VGA port because the analog side is usually the most simple. Since the projector shows only green components, it seems like something has killed the red and blue channel, which is consistent with the idea of a voltage spike rolling in through the attached video source devices. Low signal voltage video inputs are usually equipped with some kind of buffering or protection circuit to prevent damage by electrostatic discharge (ESD), e.g. when touching the open plug while changing connections. There might be a voltage-tolerant transistor buffering stage or a bunch of clamping diodes.
|VGA input circuit, top side||VGA input circuit, bottom side|
Now, a closer look at the PCB area just behind the 15-pin D-SUB connector revealed several identical SOT-23 SMD parts. From googling the marking codes A7W, I expected them to be either dual diodes or NPN BJTs. A diode makes more sense in this place since there is no collector or emitter resistor present. This pretty much makes the use as a buffer impossible. Also, the parts are denoted as “Dxx” on the board, so diodes it is. A quick measurement of the two separate diodes in each device (one from signal to VCC and one to GND) shows three of the pairs totally open and one with erratic values (~30…130 mV diode voltage), with the erratic one being coupled to the red line. This looks like a strong high voltage spike has hit the device and has been shorted by the diodes. Unfortunately, diodes should normally be short-on-fail, which means that IF they are destroyed, they usually short out instead of opening up. This is not a good sign, as is the not completely shorted one, because a high voltage component might have gone through in this case. I replaced the diodes with matching ones from my parts bin (which was quite the surprise for me, I never thought I really had those). Hot air soldering is recommended for this kind of work, and you should probably select a higher temperature to keep your soldering time short. The solder on this board is high-melting lead-free, which is really the worst for reworking, and the board has lots of ground layers.
After this step, the picture became green-blue. No red. Further oscilloscope-aided inspection along the signal trace shows that the signal arrives at the digital converter IC, but the input shows different impedance in comparison to the other two colours. So, I guess it is broken inside the IC. The TI AFM1000 is a custom made device which means there will be little chance of getting one from the usual suppliers, and I don’t want to look into the asian backchannel markets just because of a VGA port right now.
Fixing the HDMI port
Since HDMI is a high speed, low voltage differential signalling (LVDS) type bus, its input circuits are even more sensitive to ESD. It also has protection devices, which can be seen as long, thin plastic packages just behind the HDMI port (marking on PCB says “ED###”, probably for ESD diode):
All of these were unsoldered using hot air and measured outside the circuit. They usually consist of a pair of clamping diodes followed by an integrated common-mode suppression coil, but in this case there is only the diode part. Surprisingly, they all appeared okay. I left them outside for the moment, which was partly because these things are TINY! Even the weakest setting of the hot air tool blows them miles off the board. A reflow oven technique could be used for this, but I wasn’t sure if the larger devices are securely glued to the board. I had this happen to me once ages ago, and believe me, you don’t want to see your own face when the whole underside of your populated PCB goes *plunk* in the oven tray…but now, back to business.
At this point, I plugged in an HDMI source – my notebook – and again, nothing happened. After following some of the traces around on the board, I noticed, that the media player section also feeds its output into the same dual-line HDMI receiver IC (Analog Devices ADV7612). Strange, I thought I got short flashes of good picture from the media player earlier. So, the HDMI can’t be totally dead, can it? But how is it possible that the notebook does not even notice the hotplug event?
Measuring the external HDMI signal and hotplug-data lines for conductivity against ground reveals a low resistance of 5 ohms where the supposedly good port has resistances decades higher. Whoops, this looks like an internal short somewhere. To be sure, I unsoldered everything that was connected to the external HDMI signal path, just to be sure. Well, I have to admit, putting all that 0603 sized stuff back afterwards was a real pain, and the results didn’t change. This led me to an idea. Fortunately, I have access to a good thermal imaging cam with close-in focus, which I used to look at the PCB. In operation, the whole HDMI receiver region emits a LOT of heat, and especially the voltage regulators of the Analog Devices IC. I’ll add the thermal picture once I manage to find out which folder the cam has dumped it into.
As a test, I froze those with cooling spray – and just like that, the screen became dark and the media player popped up, nice and stable. I could even watch some short movies while cooling the IC with some more spray, but if it got a little bit too hot the player would crash at once. So, my guess is the following: The HV spike somehow went past the diodes and hit the input circuits of the IC. The parts connected to the external HDMI line were shorted out to ground and VCC when the internal transistors went down, and the remaining 5 ohms are probably substrate resistance and bonding wires. The chip can still work on the good HDMI line because of separate circuits, but the short increases its power consumption to the maximum available from the regulator, which then runs into temperature current limit. At this point, the picture either starts flickering if external HDMI is selected, or the media player picture starts hanging.
My first (and temporary) solution was to solder a small cooling fin made of a piece of copper sheet onto the ground connector of the regulator. It is necessary to isolate the surrounding parts with a piece of plastic sticky tape or the likes. Afterwards, the media player function can be used perfectly fine, if the cooling mode of the projector is set sufficiently high. If the regulator gets too hot, the picture immediately starts hanging.
The long-term solution requires a new IC. This one is pretty difficult to get if you don’t want to pay the price for a new projector, but RS Components has it in stock for a fair price. They even supply the type with hardware HDCP keys (ADV7612BSWZ without the -P), so I will not get into trouble when connecting an external blu-ray player if I manage to get and fix a broken one at some time in the future ;-)
Swapping the IC is again done using hot air. The thermal pad is a bit tricky, but since the IC is done for anyways, I hit it right in the middle with 500°C until the pad below melted. A reflow oven doesn’t help, unfortunately, since the components don’t seem to be glued to the PCB. After removal, the pads are first cleaned and fluxed, then the new IC is installed. Re-soldering is also a tricky job which I accomplished using a Weller WS50 fine-tipped soldering iron, flux and desoldering wick in combination. The bottom pad can be soldered using hot air, or you put on a small and flat bead of solder and press the IC down on it, while soldering it all around. Not the best style, I know, but the IC doesn’t really seem to need the cooling – which I checked afterwards.
After all this work, I half expected a crisp, clear picture to pop up when plugging in the HDMI cord. Nope. Argh! The notebook still doesn’t even notice the device. But wait a second, how does it even do this? Usually, there should be a device identification (EDID) module attached to the port that tells the source what is happening at its port. It also contains valid resolution/timing and transfer mode properties. And guess what, this one is dead, too. Seems like this line also got hit by the spike.
I cross-checked the 24C02 EEPROM (next to the ADV7612) outside the PCB with a diode tester, between ground and the data lines. A good one shows no conductivity, this one does. This is bad, because even if a new chip is easy to obtain, the data is inside the old chip – and we can’t clone it! At this point a little luck helped: The (dead) VGA port had an identical method of informing the attached master (well, as the standard dictates it). So, what would happen if we swapped the VGA EEPROM onto the HDMI port? Damage is not very likely, and in the worst case the settings are simply crap – so I just tried it. And guess what, it worked right off the spot.
Of course there is a reason for this: Both systems follow the EDID standard, although the EDID signal also contains some information about the connection between source and sink, which might be invalid after the swap. The source device on the other hand can just ignore the obvious wrong parts, depending on the driver. I dumped the information using the NVIDIA driver of my notebook and analyzed the saved dump using the Wikipedia article about the EDID format:
According to this the K330 informs the master about an analog connection. The notebook simply ignores this, but my WDTV Live box for example doesn’t. This causes a warning about bad resolution selections, and I would bet that some of the 3D features of the projector (120Hz mode in full 720p) are lost, too.
This leaves three ways out:
- Fix the old EEPROM (unlikely since dead).
- Find a good dump of the HDMI EEPROM and flash that into mine.
- Reconstruct something fitting from the standard documentation.
I decided to go with the last option, but for this, access to the EEPROM was needed. As the IC is on the underside of the main PCB, this would be tricky even when the case is open. Still, one can avoid soldering wires: Remember that the EEPROM is already connected to the external HDMI port! During operation, the connected source device would act as a I2C bus master and talks to the EEPROM, reading out the ID information. This can also work the other way around! There are tools for Windows and especially for Linux (i2c-tools) that can write to almost any I2C device that is attached to an internal or external bus, in this case the GPU I2C. An example shows how you can do that in Archlinux:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
# pacman -Sy i2c-tools ... <installing messages, etc> # modprobe i2c-dev # i2cdetect -l i2c-0 i2c Radeon i2c bit bus 0x90 I2C adapter i2c-1 i2c Radeon i2c bit bus 0x91 I2C adapter i2c-2 i2c Radeon i2c bit bus 0x92 I2C adapter i2c-3 i2c Radeon i2c bit bus 0x93 I2C adapter i2c-4 i2c Radeon i2c bit bus 0x94 I2C adapter i2c-5 i2c Radeon i2c bit bus 0x95 I2C adapter i2c-6 i2c Radeon i2c bit bus 0x96 I2C adapter i2c-7 i2c Radeon i2c bit bus 0x97 I2C adapter i2c-8 i2c card0-DP-1 I2C adapter i2c-9 smbus SMBus I801 adapter at 0400 SMBus adapter # i2cdetect 2 WARNING! This program can confuse your I2C bus, cause data loss and worse! I will probe file /dev/i2c-2. I will probe address range 0x03-0x77. Continue? [Y/n] y 0 1 2 3 4 5 6 7 8 9 a b c d e f 00: -- -- -- -- -- -- -- -- -- -- -- -- -- 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 30: -- -- -- -- -- -- -- -- -- -- 3a -- -- -- -- -- 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 50: 50 -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 70: -- -- -- -- -- -- -- --
I figured out which bus was my GPU DVI bus and which device on it the projector by trial and error. After I had this info, I read the EDID directly from the device:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
# i2cdump 2 0x50 No size specified (using byte-data access) WARNING! This program can confuse your I2C bus, cause data loss and worse! I will probe file /dev/i2c-2, address 0x50, mode byte Continue? [Y/n] y 0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef 00: 00 ff ff ff ff ff ff 00 04 72 23 12 42 b0 0a 00 ........?r#?B??. 10: 22 15 01 03 6a 00 00 78 0a d7 76 b2 4a 25 c5 24 "????..x??v?J%?$ 20: 05 51 5b 3f cf 80 31 7c 45 7c 61 7c 81 c0 95 00 ?Q[???1|E|a|???. 30: b3 00 01 01 01 01 9e 20 00 90 51 20 1f 30 48 80 ?.????? .?Q ?0H? 40: 36 00 00 00 00 00 00 1c 00 00 00 fd 00 32 78 1e 6......?...?.2x? 50: 64 0f 00 0a 20 20 20 20 20 20 00 00 00 fc 00 41 d?.? ...?.A 60: 63 65 72 20 4b 33 33 30 0a 20 20 20 00 00 00 ff cer K330? .... 70: 00 4a 43 4e 30 31 30 30 31 35 39 30 31 0a 01 18 .JCN010015901??? 80: 02 03 04 00 66 21 56 aa 51 00 1e 30 46 8f 33 00 ???.f!V?Q.?0F?3. 90: 00 00 00 00 00 1e 00 00 00 00 00 00 00 00 00 00 .....?.......... a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ab ...............? # i2cset -f 2 0x50 0x14 0x81 b WARNING! This program can confuse your I2C bus, cause data loss and worse! DANGEROUS! Writing to a serial EEPROM on a memory DIMM may render your memory USELESS and make your system UNBOOTABLE! I will write to device file /dev/i2c-2, chip address 0x50, data address 0x14, data 0x81, mode byte. Continue? [y/N] y Error: Write failed
The last command instructs the I2C driver to write a value of 0x81 into the byte address 0x14 of the device 0x50 on bus 2 (i2c-2). The value is arbitrary though, it was just a test – and it failed, because the EEPROM inside the projector was write protected as it should be. At this point, I put in a small modification to unlock the EDID memory. The picture below shows where to place the solder bridge to short the write protect pin to ground. Convenient of Acer to leave these resistor soldering pads open in the perfect spot.
|EDID EEPROM solder bridge|
Electrically, this is completely safe because the pull-up resistor of this line is at least 4,7k ohms (measured) and the transistor next to it would short the WP line to ground anyways. It can probably be controlled by the main CPU to allow for EDID updates when the firmware is flashed. After placing the bridge, I tried again:
1 2 3 4 5 6 7 8
# i2cset -f 2 0x50 0x14 0x81 b WARNING! This program can confuse your I2C bus, cause data loss and worse! DANGEROUS! Writing to a serial EEPROM on a memory DIMM may render your memory USELESS and make your system UNBOOTABLE! I will write to device file /dev/i2c-2, chip address 0x50, data address 0x14, data 0x81, mode byte. Continue? [y/N] y #
Strike! No error this time. A second inspection of the EEPROM dump revealed that the data had truly been stored, and a short call of xrandr showed a change in the device properties. Now I could close the casing and still be able to update the EDID if I need to later on. If your source device should not be able to do the programming, a simple I2C capable microprocessor and an HDMI cable adapter would also do the trick.
I still have to figure out the ideal data for the EEPROM, but this is a long-term target when problems arise. For now, I can attach any of my usual sources without problems.
If any of you people reading this owns an Acer K330 and feels in the mood to drop me a hex dump of the original HDMI connection EDID in the comments below or via email (see contact page), I would appreciate it!
FOLLOW-UP: There is a copy of the K130 service manual available in the net which describes a procedure called “EDID key-in”. The correct factory procedure adapts a serial connection from the programming PC to the levels used in the HDMI/VGA DDC signals, which is ultimately equivalent to the procedure described above.
The more interesting part is about unlocking of the EDID write mode. As the manual states, hold the power key, then plug in power. Hold the key until it flashes blue rapidly (not red-blue-alternating, takes a second or two) then release it. The key will flash blue-blue-red in sequence. This is what I already suspected from the write protect lines of both EEPROMs leading to the core IC. I cannot test this because mine is already hard-patched now, but my K330 at least responds to this key combo in the way described. Since the firmware seems to be largely identical, the function should be the same as with the K130 – meaning you can reprogram the EEPROMs from the PC while the flashing of the LED continues, no soldering required. This may be a helpful trick for those trying to make their K330 fully 3D-compatible – there seems to be a lot of problems going on. Probably voids the warranty though. ;-)
One nice thing about rewriting the EDID is that the projector (or any unlocked EDID-capable device) cannot be bricked beyond recovery. I2C detection on the DDC bus works independently from display device recognition. If you probe the bus, the EEPROM will always be there waiting for commands, even if the contents are mostly corrupt.
I also got hold of a copy of the good HDMI EDID which will hopefully make my old Radeon graphics card recognize the projector. So far, it refuses to even detect its presence. Probably related to some bad EDID bits, but this makes the HD5770 the only source device in my reach that just plain rejects the projector without any debug info. Not a feature, AMD, not at all.
µ ~ 2014-12-29
It worked. The VGA EDID seems to be missing several extended descriptors necessary for HDMI and also some basic settings. Still, funny that only the Radeon noticed…
Audio through HDMI also works like a charm now.
µ ~ 2014-12-31
Forgot to mention which K330 HDMI EDID I used for the restoration. Click here to get the contents and here to see where I got it from. Note that I removed the serial number of the device from the file and replaced it with “– — — –“. Insert your own here, please. Wiki tells you how, or you extract it from your own device. Also keep in mind that the “real” S/N is stored elsewhere, so forging it won’t do ;-)
µ ~ 2015-10-26
Last but not least I checked the configuration of the projector through its service menu. To enter, press the sequence POWER, LEFT, LEFT, MENU either on the device itself or the remote – makes no difference. All further instructions are given in the menu. As far as I could see, there are not a lot of options that can brick the projector, except for the LED current programming which is better left alone! Excluding this, the menu features test pattern modes, several interesting options and diagnostic tools. I checked for any options to unlock the EDID ROM before soldering the contact bridge, of course.
- Bought an Acer K330 LED DLP projector as broken for roundabout 40 Euros.
- Fixed the VGA input circuitry, but VGA is still partially broken. No red colors.
- Composite input works, audio works.
- Removed the probably damaged HDMI ESD protection circuits. No replacement so far, will install some later on, but they are a pain to solder because of their small size and contact density. Until then, being careful and always plugging the projector HDMI in last and unplugging it first should prevent static from getting to the data pins.
- Replaced the HDMI receiver IC for ~13 Euros.
- Dropped the original HDMI EDID chip and swapped the (now useless) VGA EDID IC over. Probably suboptimal data, but it works.
- Modified the EDID write protect to always-writeable to allow editing over the HDMI cable.
- Strangely, the projector does not allow audio through HDMI. That is no big problem though, since I prefer optical connections for this.
Some useful hints for checking circuits:
- Checking BJT transistors in circuit can be done by testing the two diodes with a multimeter. Remember the circuit symbol: NPN (“not pointing in“) has two from base to C and E while PNP (“pointing in proudly“) has them reversed. Typically, the forware voltages should read ~500-900 mV depending on the type, and if both do, it is probably fine. Dead ones can show up as 0-200 mV or larger than 1800 mV. The reverse voltage should always measure larger than 1800 mV, infinity would be ideal. This method works in ~90% of all cases. BUT: Take care of any parts that are mounted in parallel to the diodes, like other reverse diodes. Usually, at least the reverse voltage reading will be off and the device will need to be removed for correct measurement. Keep in mind that the diode function is only one part of the whole, a transistor showing good diodes may still be broken! If further measurements indicate that something simply does not fit, desoldering of components or well-placed cutting of signal lines to isolate function blocks is needed.
- As already mentioned, it can be useful to disconnect function blocks from each other and test them separately. This works well with audio amplifiers: You can usually AC-separate consecutive amplifier stages at their coupling capacitors without breaking the DC bias conditions inside each block. Take care with open inputs as some stages may tend to oscillate without an attached source – these inputs could for example be terminated with 10 kOhms resistors. Again, take care of possible un-isolated bias voltages at the input!
- Trying to understand circuits without a schematic by just recognizing the basic function blocks (which you see everywhere) is good training for the brain. Do this more often, and you will see from experience, when a measurement in circuit is off.
- Search for redundant circuits or sections. Remember: Designers also like to copy&paste, which is why most devices that feature multiple channels of anything will have the same circuit for all channels. This gives you a useful chance for reference measurement while, for example, diagnosing the dead channel circuit in an audio amp.
- For hard cases, there is always the internet: Try to google/search for your device type number adding “service manual” or “schematic” and see if something useful comes up. For audio stuff it almost always does, video and digital equipment is rather rare.
- Always, always, ALWAYS unplug and discharge your circuit before doing continuity/diode measurements. Aside from all readings being off when the circuit has residual power, you put yourself and your meter at risk of shock damage – which should be kept in mind for all other work on circuits, too, of course!
With this I hope to have given you some insights, or an enjoyable read at the least. See you around!