Design errors in commercial equipment

@ostripper
"The biggest factor is that the public has been "conditioned" to accept planned obsolescence and continuous consumption as the norm.
They would not know quality if it fell on their heads , or they they would be told lies like "you need to upgrade". Yeah , upgrade to
something that will have a shorter lifetime than the last item. "Scamerica" ... at it's best."


That's one of the best descriptions of current society that I've seen posted anywhere.

The other week, I was walking back from the grocery store, it was "Trash Day", and one of those Polk subs were put out on the curb among the other trash.
As I was walking towards it, a bird landed on it and crapped, then flew away as I approached.
I had to laugh.
 
I would of taken the woofer out , then the birds would have a nest. BTW , I'm sure the owner of the sub had (maybe) a year. then he could
just go buy another one for 200-400$.
Take a Bowers & Wilkins ASW608 - 700- 999$ . "Bowers and Wilkins" sounds so grand and formal.
(picture) 400W SMPS/Ice class D that fails and takes out the woofer. Nice woofer (pix).
B& W DOES make a good sub , you just have to go up to 2000$ (Hypex 1000 + Hypex SMPS) , woofer is beyond bada$$.
2K is a lot just to get to the quality level that was the norm for cheap stuff in the 20th century. I still find 80's amps made in Japan
that just need filter caps and TLC.
I'm glad I know the "insides" of this consumer fluff. I buy 7$ SMPS's on Ebay and replace all the caps with Nichicon's/Panasonic's. Tidy up the sloppy
soldering and switching devices with no thermal compound - the new (reworked) now....12$ SMPS lasts decades.
My XT-12 sub is at the level of the B & W 610 @ $899. But I built my amp for <100$ and got the whole Polk sub for 40$ (new with blown amp).
I just have a real class A/B and a 300VA toroid ...I'd rather have that than a supposed China 300W "iceamp".
 

Attachments

  • BW608.jpeg
    BW608.jpeg
    99.6 KB · Views: 154
I was informed of a youtube video from a polymer consulting firm in an area of interest to our company. Looking at more of their postings I was intrigued to see an entire presentation regarding the failure of plastic nuts replacing metal! If you don't follow the instructions precisely for tightening the nut, it will fail 100% of the time.

This is really important as companies seek to "light-weight" vehicles substituting plastics for steel.
 
Here's two more.
The first is an infrared thermometer I got from Costco years ago. Works great on Celsius but on Farhenheit, at temps below 32F, it has math errors. The photo is a reading of my freezer. No wonder the ice cream is always soft.
The second is the control panel for a humidification system we have at work. After 10 years the display died, so they sent us a replacement. This worked but the text was backwards. After a year, three more replacements, two firmare updates and countless phone calls, we finally had a system that could be read without a mirror.
 

Attachments

  • IR Thermometer.JPG
    IR Thermometer.JPG
    263.3 KB · Views: 107
  • driSteem.JPG
    driSteem.JPG
    241.9 KB · Views: 100
For what type of battery? It would be perfectly normal for a standard charger for NiMH batteries, a sure way to kill the batteries for a fast charger for NiMH batteries and a major fire (and possibly explosion) hazard for lithium-ion batteries.
 
A while back someone posted about burned parts in a Krell KAV300i amp. I recently bought one in working condition on ebay, assuming I would need to recap it. When it arrived I opened it up and sure enough, the main power supply filter caps were bulging a bit. In a previous post someone posted pix of some toasted parts in their amp. Mine had the same issue.

In my amp, two resistors in the 18V regulator circuit were cooked so badly that the outer cover of the resistors had turned brown and cracked. The two 10uF 50V caps between the transistors had also turned brown:

power amp board top toasted.jpg





burned resistors.jpg


The VDC references the +/- 60V main power rails, so the 6.19k resistors should only have about 7 mA going through them which works out to 0.28W, so they are not likely to be the main heat source. The transistors have 42V from C to E. If there's even 50 mA there, they will be dissipating about 2W each, so yeah, they're probably going to run warm.

The transistors have isolated tabs, so it should be easy to attach a heatsink even in the limited space between the poweramp and preamp boards.

I have replaced all the electrolytic caps in the amp- Nichicon EKWs in most spots, and for the 10uF 50V caps I used some Kemet ultra long life, low ESR parts. I'm waiting for a few more parts to arrive then I'll do the final work on the amp.
 
With respect to the Krell KAV300i, is the transformer center-tapped? I know that's a silly question but often folks try to create a synthetic ground rail without taking into account the differential current drawn by the poz and neg rails over time.

The service manual I have is missing the main power supply page(s), but there are 4 wires connecting the transformer to the power amp board, so it looks like two windings, each with a bridge rectifier. They're probably tying two of the rectifier outputs together on the DC side.

Maybe if the 10uF capacitors C21 and C46 get more leaky with time & heat, the current through the zener resistors might increase more than the original designer intended.
That seems possible, but the toasted area is centered on the two transistors, with heat conducted up and down (in the photo) by the PCB traces. I'll be adding small aluminum strips as heatsinks to the transistors and standing the new resistors off the PCB. In the photo the resistors connect at the circled pads. I should probably go ahead and replace those transistors as they have apparently been running hot for a long time. Those two parts cost about $3, so an easy and cheap replacement- no excuses...
 

Attachments

Last edited:
That's the plan. I'll be replacing the resistors, electrolytic and film caps, transistors, and Zener diodes in that area of the board. It should be pretty easy, as long as the cooked traces don't lift off the board when I desolder the parts. I'll post a photo of the final fix when it's done.

The biggest flaw in this amp's design is probably the fact that they put the output transistor heatsinks inside the chassis with minimal ventilation holes. Even if I were to cut more holes in the top and bottom covers, the heatsinks are oriented horizontally with fins running from front to back, not very conducive to convective air flow, though I suppose if you give the hot air a means to escape, it will be replaced by cooler room air.

Maybe I'll mill some slots directly over the heatsinks in the top and bottom covers...
 
Last edited:
Those two small electrolytics in between the power transistors are the ones I'd be most worried about. Nichicon's UBX product line is good up to 150C , and so is Vishay Beyschlag series 160 RLA. Both are in stock at DigiKey. It's easy to find higher wattage diodes and resistors, but not electrolytic capacitors so much.
 
The other week, I was walking back from the grocery store, it was "Trash Day", and one of those Polk subs were put out on the curb among the other trash.

My neighbour tossed out their Polk sub because apparently it sounded like it was farting out the bass. Just for kicks and grins I opened it up and found the problem in a trice: the power supply filter caps were custom from Bulge-O-Matic, which wasn't unexpected since they were rated for 35 volts -- on 36 volt rails. Swapping them out for 50 volt parts sorted the thing right out and it's been running without issue at my sister's place since 2007.

It's a piece of cake fix and you can score brownie points with friends & family with free subs!
 
  • Like
Reactions: MarcelvdG