Author Topic: Bizzare PCB layouts  (Read 3766 times)

0 Members and 1 Guest are viewing this topic.

Offline electronics-whizTopic starter

  • Regular Contributor
  • *
  • Posts: 74
Bizzare PCB layouts
« on: January 05, 2015, 03:07:14 am »
I had a bunch of old Hitachi data sys Storage area network devices I was given. Tuned out to be junk, and I parted out. I wanted to reuse some of the SMPS units. Totally odd design pinout, took one apart see how to turn it on. The units are long and thin with plugs that slide n and out.

They had allot of small parts on one end, and what they ran to sometimes went one way to a part then came back about to where it started. Even going through the bard multiple times. I thought who on earth designed this. It works, but it seem seems like a silly setup.
 

Offline TriodeTiger

  • Regular Contributor
  • *
  • Posts: 199
  • Country: ca
Re: Bizzare PCB layouts
« Reply #1 on: January 05, 2015, 04:33:05 am »
Come on, photos?
"Yes, I have deliberately traded off robustness for the sake of having knobs." - Dave Jones.
 

Offline electronics-whizTopic starter

  • Regular Contributor
  • *
  • Posts: 74
Re: Bizzare PCB layouts
« Reply #2 on: January 05, 2015, 06:00:19 am »
Pictures of the PCB.
 

Offline Falcon69

  • Super Contributor
  • ***
  • Posts: 1482
  • Country: us
Re: Bizzare PCB layouts
« Reply #3 on: January 05, 2015, 06:09:32 am »
Maybe they were using the traces for resistance, instead of using actual resistors?

I don't see anything odd about the board.  Except for maybe it being long like that, but maybe it is for a reason, like the particular case it was in connected both ends of the board to it.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16284
  • Country: za
Re: Bizzare PCB layouts
« Reply #4 on: January 05, 2015, 01:14:16 pm »
Was there not a fan blowing over that area of the board, so they wanted to keep it unobstructed for better airflow, and as well they probably had too many traces for a cable loom and 2 connectors to be cheaper than plain bare smooth PCB material.
 

Offline sunnyhighway

  • Frequent Contributor
  • **
  • Posts: 276
  • Country: nl
Re: Bizzare PCB layouts
« Reply #5 on: January 05, 2015, 07:56:34 pm »
To me it looks like they lowered production-cost by:

 - getting rid of a cable loom/flat-cable and connectors.
 - reducing the number of building-blocks/assembly-stations for final assembly.
 - reducing the number of screws needed for final assembly.
 - using the pcb for structural integrity.

And as a bonus they also got rid of a primary source of defects (interconnects), which in its turn adds to:
 - less returns under warranty.
 - better brand reputation.

 

Offline Richard Crowley

  • Super Contributor
  • ***
  • Posts: 4317
  • Country: us
  • KJ7YLK
Re: Bizzare PCB layouts
« Reply #6 on: January 05, 2015, 08:11:59 pm »
That doesn't look anything like a PS board to me.  It looks more like the signal processing circuit for a hard drive or something.
I see only one mediocre-size filter capacitor (at the bottom of the photo) and no transformers (which would be expected in a SMPS).
And what are those white components in the middle of the board?  They almost look like opto-isolators or something???
Those traces you are talking about (which I don't really see in your fuzzy photos) could very well be shielding around the signal traces.
Remember that serpentine routing is also used on PC boards (see almost any PC motherboard) to equalize timing between signal paths.
 

Offline electronics-whizTopic starter

  • Regular Contributor
  • *
  • Posts: 74
Re: Bizzare PCB layouts
« Reply #7 on: January 05, 2015, 09:10:56 pm »
This is from an SMPS from a storage server array. The main SMPS was on a different board, and screwed into the two tall terminals.  The other shorter terminals linked to output of a small DC-DC converter module that plugged into the largee 4 pin header. I found odd because they had some small logic pins on the header. They went clear to the other end though the board 2-3 times in spots, and some just return to some outputs for sense or something. Another I think to turn it on went to a SMD transistor then back to the small PCB on the side that connected to the SMPS. 

It had long thin case like length, width of the PCB, and about 4 in high. Seems of for basically being a SMPS like a PC would use. It has some diagnostic lights, etc so I could see being marginally more complex, but I don't see why they made it like that. Every SMPS I have worked on output is filtered right off the transformer. This had an SMPS running like 54V output, then running DC-DC converters. This had no input for any sort of 48V battery, etc.

Was made by or for Hitachi Data Systems which I think is a japanese company.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf