HK Audio Lucas Performer 900W Sub : Sometimes Repair Jobs Go Horribly Wrong

[@Dutch_off_grid_homesteading]

Bedankt. I really like these kinds of video's just love it those rabbit hole but I really hope you'll find the fault

[@LearnElectronicsRepair]

Thank You 🙂

[@cliff_hewitt]

Love the diagnostic procedure. Fingers crossed you can figure it out for the third video in this series!

[@budgiefish]

As you've discovered, attaching non-isolated scope grounds to 'grounds' of speaker terminals does stuff like this, particularly when inputs are also grounded and effectively shorting out the 'negative' speaker connection to ground where it doesn't belong! I've chased my tail a few times with mains grounded scope on mains isolated amp (no problem until you connect an input with same ground as scope - usually my sig gen), your mixer in your case! So either do what you did, isolated battery scope, or isolated battery sig gen/audio source. I still think the issue lies on that surface mount board as I emailed you the other week - quite possibly O/C resistors!

[@LearnElectronicsRepair]

What fooled me with this one, but I should have realised sooner due to the strange layout of the main smoothing capacitors, is that there are four channels. Each one has a speaker out. You can connect any two channels as a bridge and in this case neither end of the speaker connects to ground it connects between the two amplifier outputs and is driven in full bridge mode (high and low side mosfets at both terminals of speaker) The channels can also be driven in half bridge or single ended mode, in this case one speaker terminal connects to the amplifier output and the other connects to ground. There is a 'common' terminal for the speakers connected in this manner and i expected that to be ground. This is where I connected my scope ground. However in this amplifier the 'common speaker ground' is not ground at all it goes to the junction of the two 100V bulk smoothing capacitors. Which also have another 200V capacitor in parallel with them. This is more like the half bridge arrangement in an ATX PSU, And I've not seen that before in an amplifier so I wasn't prepared for that type of circuit. At least it did no harm, possibly because my bench mains is isolated via a 220V:220V transformer anyway

[@Hellhound604]

Yep, these bloody class-D amps can be tricky, esp. if they are used in bridge-mode. You have to view them as switching power-supplies, and use the methods you use in Switch mode PSU’s (any topology) rather than what you would use for diagnosing normal class AB audio amps. That include use of differential probes, etc. .the higher their outputpower, the worse they become. I know you do it sometimes, but when I encounter something like this, I would roughly reverse engineer the thing until I have a fairly good understanding of how the circuit is supposed to work, before going deep into diagnosis. Not sure if it saves time, but I always feel more comfortable working on high-power stuff when I have a full understanding how the circuit is supposed to work. The speaker also forms part of the filter network, so you need to have a speaker or dummy load connected to the outputs to get a meaningful output waveform, as terrible as it might sound……Love you video, as always.

[@LearnElectronicsRepair]

I get that, but the four channels could be (and two are) driven in single ended mode. In this case I expected one end of the speaker to be connected to ground (like as in the centre tap on the transformer secondary which supplies the + and - rails. This is not the case in this amplifier, but hey I'm learning new stuff like the rest of you 🙂

[@Dutch-linux]

timestamp 52:35 looks like you got some serious dc on the outputs !!!!! time to take the heatsink off and go over every component on that board !!!

[@TheElectronicDilettante]

I was reading the troubleshooting guide from HK and the way it reads is like the ground disconnect button could be bad or subwoofer potentiometer

[@Karthor.]

I think the switches are correct found a french forum that someone had accidentally flipped a switch and was wondering what configuration was correct someone in there took his apart and said the first 4 was down and the last one raised

[@LearnElectronicsRepair]

Thanks for the info Karthor

[@Pleusch]

Do you have the Thile small Parameter of the Subwoofer driver ?

[@LearnElectronicsRepair]

I don't. This amplifier was repaired on a later video. kzbin.info/www/bejne/p6G8amScbLSkhLs kzbin.info/www/bejne/q3y4p3duiLWnetU

[@Dutch-linux]

why not use klip leads to connect the grouds to each other and then connect to the case that will close the ground planes !!!

[@ralphj4012]

Not sure, can't recall, whether you checked the presence of the negative supply rail. Thought that all you kit was on an isolated supply so grounds shouldn't have mattered, but that odd high frequency switching signal and large DC offset may not have played well with the scope or probe capacitances. If it was me, having reverse-engineered part of the amp, I would drop the heat sink and continue the process (a rough sketch on paper) and then disconnected the preamp and fed directly into the amp, with a dummy load, depending on circuity. As always, only my opinions. A DIP switch which can apply 100V direct to a speaker connection, would give me nightmares.

[@LearnElectronicsRepair]

I checked both power rails in this video, +/-75V give or take. The device under test is always connected to the isolated supply and a lot of my other test equipment is hand held or runs from USB type PSU and therefore also isolated. But my Rigol, my Computers and Analog scope are grounded. The DIP switch would set one amplifier in antiphase to the other allowing them to run as a bridge. It is also conceivable it may enable/disable amplifier channels as well, but not by disconnecting the supply rails

[@ralphj4012]

@@LearnElectronicsRepair Ah, it may be worth connecting the scope grounds to the chassis (BUT there seemed to be a peculiar separate PCB ground points arrangement and it wasn't clear to me where the mains earth connection went, even though that would have been effectively isolated initially). As always, clean trousers, defibrillator etc. at the ready.

[@mrkikiboy]

I need the power supply diagram. do you have it available?

[@teslasapple]

Can’t win them all, just most…and that’s fine 😊

[@melvilib]

Often you draw a circuit diagram if there is not one available so we can understand it. Would that not help you understand the circuit you are dealing with?

[@Dutch-linux]

you should check the bias if that is pressent !!!!!! or one of the mosfet's is shorted that switch should not have that kind of effect ... there is something wrong with the circuit

[@Hellhound604]

Class-D amp, so no biasing. However, it means that one of the MOSFET’s in the half-bridge is either shorted, or not permanently swiched on.

[@Dutch-linux]

@@Hellhound604 still a very bad design !!!

[@Dutch-linux]

timestamp 49:14 the amp is oscilating think by powering it up without the grounds you blew something up ... this is a very very bad design !!!

[@andrewreed7101]

Your killing is with ads

[@LearnElectronicsRepair]

How many ads do you see in this video? It seems to vary from person to person. KZbin place the ads I don't have any control over that

[@andrewreed7101]

@@LearnElectronicsRepair the last two episodes showed 2 ads every 5 minutes. That is way too many

[@brucenicoll4373]

I have to say h k is the worse brand in have ever worked with and sound so disgusting with every model