• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Squeezebox Touch PSU

Status
Not open for further replies.
Parts arrived! (1st build notes...)

Parts from Digikey arrived in less than 48 hours - amazing...😱

The BOM as supplied works well for Digikey, all parts were in stock (a blessed first for me ...) and are pretty fairly priced at about $68 per board including the Triad transformer - but excluding the mounting and chassis hardware.

Assembly is quite straight forward, if one pays attention to Jean-paul's notes and doesn't spend some minutes rabbiting about in the parts bags looking for C1 and R1 - which are for the optional snubber which I'd chosen not to install (having dealt with English wine merchants, I've been more than sufficiently snubbed already...)

Discovered one minor 'gotcha' - the X4 connector - which I believe is for the LED is not correct for the indicated board placement - but the LED wires solder nicely onto the board - do observe polarity, it's positive lead should go away from the X3 connector, closest to the X5 (ammeter) connection, and facing the large capacitor - like this:

https://www.amazon.com/clouddrive/share/ukg3iQE7Fv2j3m8eq4Fpjp9x4y3rHfVr9WiUJBWx2TP?ref_=cd_share_link_copy

I'm not sure that it's necessary to electrically isolate the LT1764 from the heatsink, it doesn't appear to be grounded, but lacking the schematic, my native paranoia won out.

One other item of note, apparently R5 and R6 are paralleled, since there's 0 ohms showing between the R6 pads and the R5 through-holes on the bare board - I can only assume that this was deliberate and not just my sloppy soldering.

https://www.amazon.com/clouddrive/share/5ofZIn5DQfcSMviwE4hx7FvZqyFdLvUHDNyIvJuTdF2?ref_=cd_share_link_copy

We'll stop dithering now, go set up to supply some smoke to the board & see where it all leaks out.

Cheers & thanks for the fun project...

Jim
 
Merciful heavens -

Succeeded in NOT frying this board...

Not due to any skill or brilliant analysis on my part, but chalk up a moral victory to the power of really good Bangladeshi cooking.

Went ot our favorite local & came home so satiated and tranquilized that I was sort of casually glancing at the BOM notes when enlightenment suddenly struck:

DOH! I installed BOTH the R3/R4 and R5/R6 pairs! :headbash:

*you can use through hole or SMD setting resistors or a combination.
Order according your preference or availability. 10K/3k24 combination for 4.98 V"


Which statement I now interpret to mean that one should install:

either
R3 (10K axial) and R5 (3.24K axial)

OR
R4 (10K SMD) and R6 (3.24K SMD)

to get 4.92VDC output

Otherwise install
R3 (10K axial) and R5 (3.16K axial)

OR
R4 (10K SMD) and R6 (3.16K SMD)

to get 5.07VDC output.

Without a schematic, I'm reduced to my rather poor logic - but this reading makes more sense than using SMD and axial components in parallel.

This particular difugelty came about because of my own tunnel-vision "let's install all the SMD devices first, then do have through-hole components for dessert. It's such a simple circuit, what could possibly go wrong?'

No big issue, R4 and R6 will go to SMD-heaven and the voltage should be correct.

Cheers

Jim
 
Parts from Digikey arrived in less than 48 hours - amazing...😱

Discovered one minor 'gotcha' - the X4 connector - which I believe is for the LED is not correct for the indicated board placement - but the LED wires solder nicely onto the board - do observe polarity, it's positive lead should go away from the X3 connector, closest to the X5 (ammeter) connection, and facing the large capacitor - like this:

Sorry, I now know why i got those sent by Subbu. You need WM4200-ND. Since i stock the Molex KK100 connectors i just used the one from my stock.

I'm not sure that it's necessary to electrically isolate the LT1764 from the heatsink, it doesn't appear to be grounded, but lacking the schematic, my native paranoia won out.

Not necessary.

As I understand it, R5,R6 is an either/or situation - axial versus SMD.
Same should apply to R3,R4.

If you want to use axial resistors, populate R3 and R5.
If you want to use SMD, populate R4 and R6.

Correct.
 
Last edited:
To all, I see this method of distribution is wrong when corrections need to be applied and having a builder proofreading would be wise too.. i will have to make a build manual. Those that already bought parts : please make sure you solder the diodes with at least 2 mm space between diode body and PCB. This to provide some cooling of the diodes.

*X4 should be a Molex KK 0.100 2 pin connector. WM4200-ND at Digikey.
 
Last edited:
Well, the blessed little fellow seems to run just fine - despite my best efforts to pooh it up.

4.92 VDC unloaded, 4.91 VDC charging my android phone, which is rated as a 1.5 ampere maximum draw, and the diodes and IC are running perhaps 2 degrees C above ambient - just perceptible warm after being under load for an hour.

In case anyone else needs to power a Raspberry Pi, here's a site with the microusb pin-out and cable color code:

Micro-USB connector pinout diagram @ pinouts.ru

Yup, the power pair are Red & Black - little did Stendhal realise that he was setting a standard ...

All in all, I believe that you and Subbu rather nailed this one Jean-Paul - with the exception of a single connector the BOM was spot-on, the board was an easy assemble, and the issue I had was purely of my own making.

BTW - for those ordering from Digikey, the heatsink I used was an Assmann V8813Y, Digikey part # AE10769-ND, it fit perfectly, matches the outline on the board, and is 50mm tall.

The TRIAD transformer works very well, it's compact, utterly silent, and runs cool to the touch.

A very worthwhile small project, and one which will give even further reason to annoy the long-suffering neighbors.

Cheers & thanks

Jim
 
OK, but the Triad was not verified (yet) to work OK at 3A load. The voltage at the input of IC1 should be at least 5.5 V minimum (lower limit of ripple voltage) at 3A for right operation of IC1. As explained before we possibly are possibly on thin ice here but things might work out perfectly fine. The transformer AC voltage is so low that margins can be tight. I feel it is my duty to deliver solid engineering and make sure that a BOM part is fine in all circumstances which means worst case thinking. This while having a positive spirit of course 😉

Some questions:

1. Since you have built it as the first could you please observe input voltage on an oscilloscope preferably with a high load ?

2. Is IC1's back fully covered by the Assman (what's in a name ?) heatsink ? It seems not, but it needs to be to transport heat the best way.

3. A picture would be nice.
 
Last edited:
Just occurred to me that a "Building the SBT PSU" thread would be good for the less experienced builders. There is some elementary know-how necessary to build devices safe and good. Now I am a quite busy chap but maybe one of you is the guy to have his 15 minutes of fame.... Well, it takes a bit more time in most cases 🙂

OK, here it is. Let's continue here:

http://www.diyaudio.com/forums/power-supplies/277009-building-sbt-psu.html#post4387481
 
Last edited:
Does anyone know the Digikey part number for a replacement power connector to the Squeeze box?
I don’t want to sacrifice my original power cord and connector.
Thanks Jerry

The connector is 2.5mm ID 5.5mm OD. Something like Digikey # CP3-1001-ND or CP3-002BH-ND or Mouser 490-PP3-002B or 490-PP3-002BH.
The first one is just slightly shorter than the stock connector and the second one is a few mm longer. Both work fine.
---Gary
 
My board and transformer arrived in UK this morning. Many thanks Jean-Paul.


So, now thinking about parts for this, and remembered that there was a hint that a parts group buy may happen...

Hi JP and All

Yes I'm in for managing the pats GB.
Since beginning of 2015, shipping rates have changed and increased dramatically... Economic rates have collapsed 🙁
In a view to avoid astronomical prices, parcels must be less than 3cm thick.

Regards
Phil

...is there any news on this, i.e. might this happen at all or anytime soon, or should we all look to source our own bits?
 
Status
Not open for further replies.