MoBo Memory Power Bypass

[Sligolad]

Thanks Fran...I have placed an order!
I managed to to fit the Regulator with the original heatsink inside the Streacom case so just need to try it for a few hours and see how it copes with poor air flow.
If it works bolted to the bottom panel in the Zuma then I will leave the fitting of the regulator to the Streacom case for now.

Should have replacement Mobos midweek so when I am back in the country later in the week I will setup a test run next weekend and see how it works.

[Sligolad]

OK I have taken a different approach to the memory bypass and it works well.

A little background first, I carried this modification some time ago be heating the 2 pins on the underside with a lot of heat to get any movement on the choke with a pliers, I eventually forced it off the board as the internal copper coil really pulls away a lot of heat from the soldering iron.

It worked great just by soldering a 1.3v line on the underside to the inner solder point nearest the memory slots.

The next time I tried it on a board I did not get it to work and it took me some lost boards before I figured out the problem.

Turns out as I applied more heat to these chokes the solder was running up the coil legs and getting on to the front PCB copper traces and this was killing the board, even cleaning the solder from the front as best I could it was hard to recover the boards.

This prompted me to take a different approach where I decided to crack and break up the Choke casing from the front side and leaving the copper coil attached.

To do this I bought the cheapest small narrow nosed pliers I could find at a couple of euros knowing it would not be the strongest and I easily bent the tips so that the pliers tips became like a nut cracker and perfect for cracking and breaking up the Choke casing. Just proceed carefully with pressure built up slowly and move the pliers around until you break off most of the casing.

Might be best to seal off the surrounding area with tape to save a lot of later cleaning.

When you get most of the casing broken away you can then pull on the coil with the pliers to unfold it so that you can then snip close to the board.

After that a good clean-up of the surroundings and the protruding ends of the choke wire ends for soldering wires for connections.

I had spare chokes available from other bad boards so I soldered both so I could always reconnect the choke if needed.

It is a good way to check everything is working OK if you run in to problems but for most you will not have a spare choke so just attach a feed wire for the 1.3v - 1.5v to the inner leg.

I will start uploading pictures now and make of them what you will and I can answer questions over the coming days.

Cheers, Pearse, and good luck!!

[Sligolad]

[nige2000]

Just dawned on me
we need to try supercapacitor/ultracapacitors on this ram mod
at quick glance there's a few at 1.4 and 2.1 volts

[goon-heaven]

Thanks Pearse for releasing the brutal truth on coilobotomy.

[Sligolad]

I did something this evening that was against my gut instinct but boy was my instinct wrong!!

I took the Teddy Pardo 5v supply off the regulator for the memory and took one of the 5v lines off the Paul Hynes supply which is already supplying the OS SSD and the Music SSD and added it to the 5v in on the memory regulator.

What a difference....stunning improvement in sound, added bass, body, clarity and atmosphere of the music just all snapped together.

I do not think it is just the case of the change from Teddy Pardo supply to Paul Hynes as I think the Teddy Pardo has been sick for maybe a month but it was working away happily providing the memory voltage through the Gainwind Regulator but as soon as I switched to the Paul Hynes it was night and day!

Afraid the 2 Teddy Pardos I have for 3.3 and 5v will be going to the scrap heap as they are obviously not well after a recent scrape with the Paul Hynes supply about a month ago.

I had expected that adding the memory supply to the 5v rail from the Paul Hynes with 2 SSD's already there (which obviously would not even cause it to break a sweat) might pollute the SSD's but not a sign of that, instead the best sound I ever heard from my speakers.

Having a really good power supply to memory for me now is a must given the sound difference I heard today.
Maybe those Super capacitors Nige mentioned will be well worth a look!
Cheers, Pearse.

[nige2000]

so the teddys are toast
might take a look at them before they hit the bin

supercapacitors are expensive can get into hundreds

was half looking at these
http://www.mouser.ie/ProductDetail/Nich ... Xx1zAT8%3d

each ram stick can consume up to 7w
so its fairly hungry

some 3-4000f supercaps would be interesting to try but aint on my shopping list

[minionas]

So what amperage 1.3v ps should be enough to feed two x 4gb sticks?

[Sligolad]

so the teddys are toast
might take a look at them before they hit the bin

supercapacitors are expensive can get into hundreds

was half looking at these
http://www.mouser.ie/ProductDetail/Nich ... Xx1zAT8%3d

each ram stick can consume up to 7w
so its fairly hungry

some 3-4000f supercaps would be interesting to try but aint on my shopping list

Quite happy to hand them over to the TirNaHiFi R&D department Nige to do as you will with them, they have served me well over the years so they owe me nothing.
They never had a chance once the Paul Hynes was let loose on them....

I will leave the Supercapacitors to the R&D department to test as I expect they would be a match for the Paul Hynes and would most likely make toast of it if they were put anywhere near my hands :-)

[Sligolad]

Stuck the meter on the supply line for memory this evening during boot up (2 x Crucial Ballistix Sport Very Low Profile 8GB Single DDR3-1600 1.35V UDIMM)

3 amps during power start up and memory scan and then 5 amps for a brief moment during OS firing up and then it settles on 3 amps running.

I have these sticks running at 800Mhz and 5 - 5 - 5 - 10 timings.
Linear supply at 1.35v.
Cheers, Pearse.