Pages:
Author

Topic: GridSeed 5-chip USB miner voltage mod - page 66. (Read 156991 times)

sr. member
Activity: 378
Merit: 250
March 28, 2014, 05:19:40 PM
Hello all.

Earlier someone made an important point and I'd like to re-iterate and re-state it.

Regarding modifications and overclocking:
the only thing that matters in the end, is increased pool side accepted shares and pay out! $Moolah!$
That's why we're all trying to squeeze more hashes out of these little beauties! Mo Money!
We're not doing it just to make it run faster! 1400MHz ain't for shyt if it doesn't yield more profit much less burns the little buggers out prematurely!

There is obviously a point where overclocking no longer yields positive results at the pool, not to mention making magic black smoke Wink. Seeing positive results locally 'client side' means nothing if it isn't reflected pool side!

So when reporting your overclock results, please be sure that you have not only achieved stable operation with extremely minimum to ZERO hardware errors locally, but most importantly, improved hash rates and $increased profit$, pool side!

No increase in profits = negative results = wasted energy and of course, ones valuable time!

And so, The Evolution Continues!

Fantastic work my fellow Techies! Keep it up!

Thank you all!
Wolfey2014

sr. member
Activity: 339
Merit: 250
Vice versa is not a meal.
March 28, 2014, 05:11:36 PM
We gonna beat you all! Kiss


BEAT THIS, GUYS Smiley




1350mhz, @31Watt, longterm proof comming soon.


wow....its just getting better, did you meant 1250mhz?
Yeah youre right Smiley.
hero member
Activity: 672
Merit: 500
March 28, 2014, 05:11:22 PM
wow 30 watts! lol intense

sr. member
Activity: 378
Merit: 250
March 28, 2014, 05:02:17 PM
Awesome! I'll gonna try to beat that Smiley!

My opinion is it's not a good idea, because
- every 100mV gives more less extra frequency and efficiency will be very bad.
- i tried to add more 280mV and it caused temperature runaway in idle state.


Solder bridges + resistor + pencil ?

Tell me your secret!

Nope, you need to change resistors only. Replace R52 with 100k, and place 120k to C32. In parallel it is 54.54kOhm, and VDD will be 1.97V.
If you would like to use only one resistor there is a 54.9kOhm in E96 series, but VDD will be more than 1.97V maybe 2V.

If it is working pls. for you, pls support my risky experiments:)
BTC: 1gqdzx8iSwUGt3vaoEaPCjvrWo7zKn7PK


I am giving out 0.6BTC to the first to get his Gridseed miner stable at 1100 MHz (<10 HW error in 24h) and post the steps to mod the miner. So far I have managed to get it stable at 1013 MHz, but I feel we can push it further.

Disclaimer: as always, I am not responsible for damage to your miner, do this at your own risk!

So i won 0.6BTC  Shocked

So, you're saying to put a 120k resistor across C32 in parallel with it, right?

If one chooses the single resistor method of 54.54k instead, it is to replace R52 instead, right?

No longer any need for the 4 previous mods i.e. 2 jumpers, 39k and pencil mod, correct?

The ultimate option / mod is:
Just a single resistor (54.54k) solution and that's it?

If so,,,,
How sweet it is!  Grin

Wolfey2014
legendary
Activity: 1722
Merit: 1000
Satoshi is rolling in his grave. #bitcoin
March 28, 2014, 04:47:45 PM
ok, so i did the mod from original post, 1 bridge only, i can get to 938 mhz with very low amount of hw errors.. cca 20 in 24hours.
pools also confirm 400 avg. on 24 hour run., tested on ghash.io, 15min,1hour,and 24hour say the same, 2mh for 5 units moded.

i see that u guys pushed further, can anyone confirm what more mods EXACTLY is needed to be done to get past 400?

and what do you have extra in your .bat? i only have added voltage=1, should there be more ?

thanks and realy good work guys !










.
hero member
Activity: 826
Merit: 1000
°^°
March 28, 2014, 04:32:07 PM
we should do a table with frequency vs hashrate vs wattage

to find the sweet spot of efficiency and maximum with 5V fan
legendary
Activity: 1855
Merit: 1016
March 28, 2014, 04:21:55 PM
Suddenly my wiibox controller page showing new thing "Automatic mining"





member
Activity: 71
Merit: 10
March 28, 2014, 04:20:36 PM
We gonna beat you all! Kiss


BEAT THIS, GUYS Smiley




1350mhz, @31Watt, longterm proof comming soon.


wow....its just getting better, did you meant 1250mhz?
sr. member
Activity: 462
Merit: 250
March 28, 2014, 04:13:28 PM
damn 532. What in the hell
sr. member
Activity: 339
Merit: 250
Vice versa is not a meal.
March 28, 2014, 03:57:16 PM
We gonna beat you all! Kiss


BEAT THIS, GUYS Smiley




1250mhz, @31Watt, longterm proof comming soon.
SVK
sr. member
Activity: 378
Merit: 250
March 28, 2014, 03:47:06 PM
Wouldn't it be better to measure submitted shares ?

GDS can hash like crazy but submitted shares are the ones which generate you coins.
sr. member
Activity: 378
Merit: 250
March 28, 2014, 03:42:24 PM
I'm seeing much better results at my pool on this latest trial run.
24Hr of 444KH/s to 575KH/s Wink

575 on average or "speed" ? BEcause 575 as average would be mindblowing and the same as claiminig your last report as 150%.



Maybe we should start to only post of numbers that are correct instead of peaks? like 24hours-average or cgminer reports. Really, this is confusing and people think you reached numbers you didnt

I agree but.....
Found out that ltcpool stores avg data via cookies and when I cleared them the actual avg showed up., as above.
I had previous non modded run still in cache so it was not showing the actual 'real' average like it should have been.
There is no intent to try to fool anyone. In fact the evidence speaks loudly of this!
I have re-started the test but based on what I've learned recently, I am leaning more and more toward changing over to cgminer from cpuminer. Too much guess work otherwise and as you can see, it pulls in haters, whiners and doubters too much. Poor things.  Cry I don't need this shyt! So on with a better means of finding the actual maximum stable $profit producing$ sweet spot!
Wolfey2014
hero member
Activity: 826
Merit: 1000
°^°
member
Activity: 107
Merit: 13
March 28, 2014, 02:58:02 PM
Nope, you need to change resistors only. Replace R52 with 100k, and place 120k to C32. In parallel it is 54.54kOhm, and VDD will be 1.97V.
If you would like to use only one resistor there is a 54.9kOhm in E96 series, but VDD will be more than 1.97V maybe 2V.

Wow, 2V DVDD is impressive - thanks for taking the risk Smiley

I have achieved 1401mV DVDD and 1204mV VDD_PLL (measured) with this approach:

- Enable VID1: Short R336 (original VMOD)
- Increase DVDD to 1.4V: Cut R46 (0R jumper) and replace with 2.94k
- Increase VDD_PLL to 1.2V: Add 27.4k on top of R212 (27.4k || 10k = 7326R)

For reference, the default VDD_PLL is 0.8 x (R211 + R212) / R212, or 1.092V.

Not getting HW errors at 1000 MHz, but I haven't tested long term because I only got my units yesterday. I'm sure 1100+ could be achieved by increasing DVDD using the above method.



You dont need to get 0 HW. 1187.5/1200-> 98.954%, so if you get less than 1.041% HW error that is enough at 1200MHz.


BTC: 1gqdzx8iSwUGt3vaoEaPCjvrWo7zKn7PK
hero member
Activity: 525
Merit: 500
..yeah
March 28, 2014, 02:50:31 PM
I'm seeing much better results at my pool on this latest trial run.
24Hr of 444KH/s to 575KH/s Wink

575 on average or "speed" ? BEcause 575 as average would be mindblowing and the same as claiminig your last report as 150%.



Maybe we should start to only post of numbers that are correct instead of peaks? like 24hours-average or cgminer reports. Really, this is confusing and people think you reached numbers you didnt
newbie
Activity: 20
Merit: 0
March 28, 2014, 02:40:48 PM
Nope, you need to change resistors only. Replace R52 with 100k, and place 120k to C32. In parallel it is 54.54kOhm, and VDD will be 1.97V.
If you would like to use only one resistor there is a 54.9kOhm in E96 series, but VDD will be more than 1.97V maybe 2V.

Wow, 2V DVDD is impressive - thanks for taking the risk Smiley

I have achieved 1401mV DVDD and 1204mV VDD_PLL (measured) with this approach:

- Enable VID1: Short R336 (original VMOD)
- Increase DVDD to 1.4V: Cut R46 (0R jumper) and replace with 2.94k
- Increase VDD_PLL to 1.2V: Add 27.4k on top of R212 (27.4k || 10k = 7326R)

For reference, the default VDD_PLL is 0.8 x (R211 + R212) / R212, or 1.092V.

Not getting HW errors at 1000 MHz, but I haven't tested long term because I only got my units yesterday. I'm sure 1100+ could be achieved by increasing DVDD using the above method.

member
Activity: 107
Merit: 13
March 28, 2014, 02:40:15 PM
Sorry to disappoint, but that's not yet stable by my standards, which is less than 10 HW error in 24h. But obviously the highest stable frequency wins, so it will be you if you manage to get it stable...
Now that, folks, is called pushing it to the limit, great work mate.  Wink

It is stable at 1100MHz. I can prove it if you need...
hero member
Activity: 616
Merit: 500
March 28, 2014, 02:32:49 PM
Awesome! I'll gonna try to beat that Smiley!

My opinion is it's not a good idea, because
- every 100mV gives more less extra frequency and efficiency will be very bad.
- i tried to add more 280mV and it caused temperature runaway in idle state.


Solder bridges + resistor + pencil ?

Tell me your secret!

Nope, you need to change resistors only. Replace R52 with 100k, and place 120k to C32. In parallel it is 54.54kOhm, and VDD will be 1.97V.
If you would like to use only one resistor there is a 54.9kOhm in E96 series, but VDD will be more than 1.97V maybe 2V.

If it is working pls. for you, pls support my risky experiments:)
BTC: 1gqdzx8iSwUGt3vaoEaPCjvrWo7zKn7PK


I am giving out 0.6BTC to the first to get his Gridseed miner stable at 1100 MHz (<10 HW error in 24h) and post the steps to mod the miner. So far I have managed to get it stable at 1013 MHz, but I feel we can push it further.

Disclaimer: as always, I am not responsible for damage to your miner, do this at your own risk!

So i won 0.6BTC  Shocked

Sorry to disappoint, but that's not yet stable by my standards, which is less than 10 HW error in 24h. But obviously the highest stable frequency wins, so it will be you if you manage to get it stable...
Now that, folks, is called pushing it to the limit, great work mate.  Wink
member
Activity: 107
Merit: 13
March 28, 2014, 02:04:51 PM
Awesome! I'll gonna try to beat that Smiley!

My opinion is it's not a good idea, because
- every 100mV gives more less extra frequency and efficiency will be very bad.
- i tried to add more 280mV and it caused temperature runaway in idle state.


Solder bridges + resistor + pencil ?

Tell me your secret!

Nope, you need to change resistors only. Replace R52 with 100k, and place 120k to C32. In parallel it is 54.54kOhm, and VDD will be 1.97V.
If you would like to use only one resistor there is a 54.9kOhm in E96 series, but VDD will be more than 1.97V maybe 2V.

If it is working pls. for you, pls support my risky experiments:)
BTC: 1gqdzx8iSwUGt3vaoEaPCjvrWo7zKn7PK


I am giving out 0.6BTC to the first to get his Gridseed miner stable at 1100 MHz (<10 HW error in 24h) and post the steps to mod the miner. So far I have managed to get it stable at 1013 MHz, but I feel we can push it further.

Disclaimer: as always, I am not responsible for damage to your miner, do this at your own risk!

So i won 0.6BTC  Shocked
newbie
Activity: 16
Merit: 0
March 28, 2014, 02:04:07 PM
so anyone tried swopping out the oscillator crystal (x6 tink its marked no?) for a higher frequency one?

alos whats wiht the reistor values, i mean where did the 39k reisitor value come from? could it be higher?Huh (in line wiht chip u9 upo1111)

and laslty has anyone got or know where to get hold of the data sheet for the 2-bit VID buck converter chip u2, (up1509)

and for ref r211 is labeled a 3.65k -1% 04 and R212 10k 1% 04

And final why do the solder bridge to short out vid0 and vid1 ? then replace the matching reistor. Why not just replace the corresponding resistor to vid0 open and vid1 open? which i believe would be R139 default value set at about 22k. )

or did i miss something.

i try but i dont have the values and compents sizes -at the moment - to try.

Are you an electronics engineer? I'm interested in any hack that works well and is the easiest one to do.
I found one other shortcut that negates the need for the pencil mod part. I'm still testing its results.
So far, 135% to 150% stable gain in overall hash rate has been achieved at 1000MHz.
1100MHz throws less red nonce HW errors but it's still too much.
1200MHz stable is still pretty much impossible at this time with current mods.
PM me if you'd like to correspond and we can figure something else out together.
I wish I knew where Sandor went off to... He started this whole thing anyway Wink
Wolfey2014


thanks ill take that as compliement - electronics engineer - but no. i did study appplied physics wiht computer over 12 years ago. didnt really get into it as a carreer started travelling instead. now ive forgot more than i know. still know the basics and beyond and dabble in it as a hobby - heck been into the electronics and computer since a kid wiht my dad. assembled a circuit at 5 and wrote a compuiter program on spectrum 48k at 7. ahh back in the days. problem i now have is though that i live in finland and dont know where tyo get most of the stuff from - language also an issue - so i result to online purcahses most of time which takes time to delivery. heck if i had the parts the crystal would be swopped out now along wiht the reistors. heck im itching to use my new rework station. oh and have things moved forward so much in 12 years. so im a bit behind. but i catch up well - just wiht coding you know the basics you can pick it up and understand it. so anyone know a good site where i good hunt for the datasheets for the chips?

Do you live in Helsinki or close by? In Helsinki (or you can just order online) there's at least this shop for electronics that I know of: http://www.partco.biz/verkkokauppa/index.php?language=en .

As soon as someone finds a way to replace the pencil mod with resistor or something (and/or when wolfey2014 reveals the new, better mod), I think I'll try to mod one of my gridseeds as well (and then the rest if I don't brick that one..Smiley

thanks ill chcek it out, especially postage - helsinki prices are pricey Smiley - not far from there about 1.5 hour on motorway. Id say youd nearby yourself? If youd like someone to mod them for you, ive got the kit, not the parts let me know.

Yeah, I live in Helsinki. I'll wait still a bit and see how this goes..  And thanks for the offer, I could see first if I can do it myself, and if not, I'm going to contact you then!:)
Pages:
Jump to: