The iSub Hack

 

This hack/mod may be a bit scant on some details and is meant for dedicated hack/makers with a bit of skill under their belt. No details are given for the physical dismantling of the iSub but, screws are under the feet (carefully remove feet) and just follow your nose after that. Use lots of padding – you don’t want to put scratches all over it as you work. All the following is used at your peril or elation 😉

I picked this up from Ebay as a purposeful hack project. Apple no longer supports the iSub, although on my Windows laptop it does appear as an audio device and I can play sound through it.

The iSub by Harmen Kardon was designed for Apple as a USB only subwoofer. It is a fairly iconic looking subwoofer, which has since been replaced with the HK Soundsticks range, using the same looking subwoofer but with two small stereo speaker stacks, all running via standard audio (USB has gone). The electronics in the iSub consists of two parts, one is the USB to audio section and the other is a plain and simple subwoofer amplifier.

The iSub is well constructed, with audio seals everywhere, even on the power socket. The USB had to go and to maintain the air seal integrity,  the USB cable would be re-purposed as the audio cable, re-wired for audio in via a 3.5mm stereo plug.

As far as reverse engineering went, I did not go as far as to re-create the full circuit diagram. It appears that the circuit board is multi-layered, so recreating the circuit would be a nightmare! I have found that the electronics in the iSub consists of two parts, one is the USB to audio section (UDA1321, USB to Audio IC + 8582C 2kbit – 256x8bit i2c EEPROM to store the audio settings for the 1321 chip) and the other is a plain and simple  amplifier (TDA7256, 30W Amp IC).

Here’s the PCB.

So what I have done…

Soldered a short across the C-E of Q01 – this stops the iSub from being constantly muted,  Q01 is actively driven by the USB audio chip, which I disabled by removing the little 3.3V regulator (top right in the picture above).

Then the audio trace from the USB audio circuitry was cut. The direct audio feed will be soldered to the cut track on the left.

Now, using the USB cable to feed the audio. As mentioned, this keeps the air-tight integrity of the iSub and saves drilling holes, etc to feed another cable. The cable diameter is a little large, but I was able to force fit a metal (for strength) 3.5mm stereo jack to the end, in place of the USB connector. The earth/ground is connected to the shield an I chose two random conductors for the L&R audio signals. For a Sub, though, you can just get away with using the Tip conductor of the plug, generally the very low frequency signals in stereo are pretty much the same.

On the PCB, I removed the small inductor array (L01) then fitted two 10k-ohm resistors to combine the L&R signals into one and then wired it to the left of the cut track.

It was then a case of a quick bench test, then re-assembling the iSub.

Final observations: The whole mod works well as an audio amplifier. BUT,  I thought that the power amp circuitry may have had some Subwoofer filtering (given the number of electrolytic capacitors around the audio path), it appears not. The subwoofer profiling may have been done on the USB audio chip, or in the Mac OS when it was able to drive the iSub, so after the mod the iSub needs to be driven by a dedicated Subwoofer output from a audio processor or a sound card.

Have fun!

Safety tip : A good supply of ice helps with soldering iron burns.

POWERTECH MI5160 180W PURE SINEWAVE INVERTER

Be Careful – Lethal Voltages inside

–   This information should only be used be qualified and technically competent  souls –

2/Jan/11 Power Tech MI5160 180W Sine wave Inverter picked up from the trash and treasure market, supposedly working (why I didn’t plug it into the car before leaving? Grrr). Attempts to power up greeted by merrily chirping of the buzzer and flashing of the Red Protection LED. So I stripped down the inverter, here’s the basic circuit layout…

and here are the internals…

What goes on in here?

The DC comes in, it is chopped up by a couple of low voltage/high current FETs (IRFZ44V) controlled by the TL494 SMPS IC, then into the step up transformer The secondary is then rectified by four high speed 3A diodes (HER308 ) and smoothed by a 68uF/450VDC capacitor. This HV DC is then supplied to, what is essentially a high voltage amplifier. A small daughter board in the unit generates the 50Hz sine wave, which is then fed to the four High voltage/low(ish) current FETs (IRF830), which in turn feeds the 240V output socket. I do not know what the two trimpots do, probably voltage settings, etc There is a heap of detail left out of this description, but as I don’t have a circuit diagram….

Diagnostics:

First thing to do was check the 12V lead, 25A fuse and DC input circuitry – all appeared OK. Then knocked the container with screws onto the floor. Next was the PCB, I did a basic power semiconductor test, and somehow missed one of the HV Rectifier diodes as conducting both ways. Promptly went on a joyride of examining each of the MOSFETs, a leading up the garden path with ESR checks of capacitors and re-soldering any suspect looking joints.

Eventually went around in a circle and got back to double checking (I thought) the diodes. Well, diode #1  appeared to be the problem, but I had no suitably fast 3A/1000V rectifier diode.

The fix:

I then stuck a 1N5408 in as a replacement (same electrical specs but not very fast at all), which proved the inverter would idle happily, but it promptly went to diode heaven after a few minutes of driving a 60W incandescent light bulb.

Farnell (Element14 or whatever) have an order for replacement diodes and FETs (just in case). Be aware that the circuit board copper is not the best, and lifts after a few solder applications.

5/Jan/11 OK, parts arrived and I replaced the HER308, and that’s all the problem appears to be. Ran the 60W/240V lightbulb for half and hour without any issues so far.

 

Fixing the Logitech S125i MiniDock

I picked up this little dock a while back and it has been a good little performer for charging and casual listening of Apple iDevices. Last night it had a sort of pink fit, turning on and off multiple times, then dead.

After checking the obvious; power supply and several different devices, it was time to disassemble and see what was going on. First trick was how to dismantle the little sucker… I unscrewed the battery holder underneath, moved it aside, and saw that the case was held together with screws from under the speaker grill area.

As you can see the white plastic surround is gently prized off, it is held on by plastic weld posts, and some of these were lost to the cause! The grill is next, again gently prying off with a sharp tool. Now remove four screws holding the speakers and the other four case screws, then pull apart.

It turns out the problem was cracked solder joints on the DC connector (circled below), re-soldering them with a little scrap wire on the PCB to give some mechanical strength, should make it last quite a bit longer.

Here’s another shot, this is the PCB…

and here we are all working once more, listening to some classic music!