Page 7 of 11
Re: MoBo Memory Power Bypass
Posted: Tue May 05, 2015 10:47 pm
by rickmcinnis
nige,
Jack used NiMH batteries for memory. Back then we were using only 256mB of memory which did not require much power. I guess what we are using now is very different in its power requirements. Though you can get some some powerful NiMH batteries. I plan on giving those a try float charged to 1.3. Brian Lowe/BELLESON is making me some low voltage regs for this.
I am not using computer power for WAVE IO which is why I wish the JCat REF cable was available as a single length.
I will be using the trusty A123s to power the USB card.
JCat says the drivers are somehow built in. I assume the drivers on their website are different - for applications where the built in drivers do not work. Don't know how I could get those built in drivers from them and if I could I could not install them the usual way. Hoping for this to be easy. Of course, this is asking for disappointment.
My OS is down to 600mB so it will not install anything. I hope these built in drivers are recognized so I do not have to start over again.
From what I am hearing there is worth is trimming down WINDOWS. When the mood strikes let me know and I can send you a list of retained files and some tricks.
So the USB recipe is a proprietary secret? I was planning on making cables from my cache of Allen Wright's silver foil until I saw that thing you are using and PANG USB which made me think I was not thinking correctly on USB transmission. I have always assumed than a GOOD single length of wire would be best but I guess it is not?
Re: MoBo Memory Power Bypass
Posted: Tue May 05, 2015 11:07 pm
by nige2000
jcat driver link is at the bottom of this page there is a performance increase over the std windows driver
you can use this driver on the silverstone either its the same chipset family :)
will the bellesons have enough current?
nimh is a good choice once theres enough capacity
my best usb cables are less than 300mm long use pure silver wire for d+ d- and gnd which is spiraled around 10mm cotton rope or teflon then shielded at the pc side
Re: MoBo Memory Power Bypass
Posted: Wed May 06, 2015 3:56 pm
by goon-heaven
nige2000 wrote:well you could place a switch between green (power on) and gnd on your smps atx supply to manually power up the psu
then when over 1.3v or so has been reached on the ram turn pc on
Rain forced me indoors..
tried your suggestion.. PS powers on, but no voltage appears across RAM cap until pc itself is turned on.
SMPS: Seasonic X460
MB: Gigabyte H61M-DS2
Re: MoBo Memory Power Bypass
Posted: Wed May 06, 2015 4:14 pm
by nige2000
Raining mad here too
Check there's power on the 3.3v and 5 v rails when PC is powered off and ps is on
Re: MoBo Memory Power Bypass
Posted: Wed May 06, 2015 5:52 pm
by rickmcinnis
nige,
BELLESONs can pass over two amps. I think that should be enough. I am using one stick of 8gB running at 1.35v at the moment.
I am hoping Brian can get a reg just below 1.3. I need to see if I can set to 1.25 v.
I was surprised at how the capacity of NiMH batteries has increased since I last looked.
http://www.batteryspace.com/msize---20ah26ah.aspx
Wish I could use such a short USB cable. I will have to use 1500 cm which is longer than it should be but I have no choice.
I know where the link for the drivers is on the page. The are saying no driver installation is required for WINDOWS other than XP and maybe vista, cannot remember.
PANG's connectors are really nice - what are you using for your cables? Of course, I would prefer to solder the cables and get rid of the connectors entirely.
Any opinion of the suitability of the silver foil for USB?
Re: MoBo Memory Power Bypass
Posted: Wed May 06, 2015 6:44 pm
by goon-heaven
nige2000 wrote:Check there's power on the 3.3v and 5 v rails when PC is powered off and ps is on
With PS on and PC off: 12v, 3.3v & 5v present on underside of 24pin connector, zilch across RAM cap.
But if RAM voltage is adjustable in BIOS, then why would voltage be present if BIOS is down?
UPDATE: The Audiowind reg has arrived in post, so hopefully no matter about above.
Re: MoBo Memory Power Bypass
Posted: Wed May 06, 2015 7:32 pm
by nige2000
goon-heaven wrote:nige2000 wrote:Check there's power on the 3.3v and 5 v rails when PC is powered off and ps is on
With PS on and PC off: 12v, 3.3v & 5v present on underside of 24pin connector, zilch across RAM cap.
But if RAM voltage is adjustable in BIOS, then why would voltage be present if BIOS is down?
UPDATE: The Audiowind reg has arrived in post, so hopefully no matter about above.
well on the up side looks like your mobo is picoless compatible :)
Re: MoBo Memory Power Bypass
Posted: Wed May 06, 2015 9:44 pm
by goon-heaven
nige2000 wrote:
well on the up side looks like your mobo is picoless compatible :)
Well, yes, I guess it is... lets hope its BendyBrokeMyChoke compatible too :)
Re: MoBo Memory Power Bypass
Posted: Wed May 06, 2015 10:07 pm
by nige2000
goon-heaven wrote:nige2000 wrote:
well on the up side looks like your mobo is picoless compatible :)
Well, yes, I guess it is... lets hope its BendyBrokeMyChoke compatible too :)
molegrips time :)
Re: MoBo Memory Power Bypass
Posted: Tue Jul 14, 2015 3:33 pm
by rickmcinnis
After my weird experience with my second BELLESON reg I am looking at the supercap idea until I muster the courage to try another regulator.
Reading about the fellow who had trouble with the supercap make me think that using a smaller polymer cap in parallel with the supercap might help - approx 1000 uf is what I have on hand and figure this will charge up quickly and the other one can take its time.
I have placed these already on the MB and think I hear a difference - just from the polymers. The ASROCK came with standard elelctrolytics.
With this board and many others, I see a similarly rated cap on the other side of the memory slots and figure it must be involved with the memory, also, so I replaced this with a polymer, too. Wonder if one should place a supercap in this position, also?
On a parallel subject - has anyone ever replaced caps around the CPU power circuit? Are those filters tuned? I would think it impossible to have a tuned circuit with electrolytics but I wonder if greater capacitance and quality there could cause more trouble than good. Any opinions?