Pages:
Author

Topic: GekkoScience Compac BM1384 Stickminer Official Support Thread - page 51. (Read 268030 times)

legendary
Activity: 3892
Merit: 4331
Does it initialize but the white light never once flashes? If so, it's a bad connection between the ASIC and PCB.

It's funny because for a short while it hashed out of the box.  Then I started running into this.  But yes at the moment it does initialize (light up) but then never a single flash.

I need to do further testing - I got the same error no matter which port on the hub I used, nor a usb port on the server.

More testing to come.  I'm going to knock the voltage and frequency way back and see if things play nicer.

Second stick from this batch is acting up now.  

 [2016-02-21 13:09:16.037] COMPAC 2: No valid hashes for over 10 secs, attempting to reset
 [2016-02-21 13:09:16.217] COMPAC 2 attempted reset got err:(0) LIBUSB_SUCCESS

BUT - again i haven't had the time to rule out the superpbag so I'm not making any conclusions.  I just need time to test.

Buddy picked up three of these I believe and he's having similar issues with two of them.  He's doing similar testing (might be issues with the superpbag).

Maybe that particular Superbpag. I have three sticks on 325 and three sticks on 275 on S-pag and they are all stable for 32 days and counting.
epg
member
Activity: 63
Merit: 10
Alrighty, well yours is the new version without a 0.9V LDO. The two resistors right above it (331 and 471) are a voltage divider that gets your 0.9V line. I don't know why you'd see 0.48V on 0.9V (after testing some 400 current-version sticks I've never seen that error) but it could be a problem there. If those resistors are seated correctly and doing their job, it must be something broken inside the ASIC.

Thanks for looking into it and responding quickly. I checked the 0.9V connection on the other 3 sticks and they show between 0.90-0.93 V, so this one does seem to be the odd stick out. I'll PM you for further steps to take for repair/replacement.
sr. member
Activity: 294
Merit: 250
Does it initialize but the white light never once flashes? If so, it's a bad connection between the ASIC and PCB.

It's funny because for a short while it hashed out of the box.  Then I started running into this.  But yes at the moment it does initialize (light up) but then never a single flash.

I need to do further testing - I got the same error no matter which port on the hub I used, nor a usb port on the server.

More testing to come.  I'm going to knock the voltage and frequency way back and see if things play nicer.

Second stick from this batch is acting up now.  

 [2016-02-21 13:09:16.037] COMPAC 2: No valid hashes for over 10 secs, attempting to reset
 [2016-02-21 13:09:16.217] COMPAC 2 attempted reset got err:(0) LIBUSB_SUCCESS

BUT - again i haven't had the time to rule out the superpbag so I'm not making any conclusions.  I just need time to test.

Buddy picked up three of these I believe and he's having similar issues with two of them.  He's doing similar testing (might be issues with the superpbag).
legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
Alrighty, well yours is the new version without a 0.9V LDO. The two resistors right above it (331 and 471) are a voltage divider that gets your 0.9V line. I don't know why you'd see 0.48V on 0.9V (after testing some 400 current-version sticks I've never seen that error) but it could be a problem there. If those resistors are seated correctly and doing their job, it must be something broken inside the ASIC.
epg
member
Activity: 63
Merit: 10
So your 0.9V is reading at 0.5V? That's very not right. Poke around on the legs of the SOT23 LDOs over on the side there and see what you get.

I Google'd what a SOT23 LDO was and there was only one on the side that I saw. Let me know if I measured the wrong points or if this is completely off. These were my readings:

legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
So your 0.9V is reading at 0.5V? That's very not right. Poke around on the legs of the SOT23 LDOs over on the side there and see what you get.
epg
member
Activity: 63
Merit: 10
Thanks for the suggestion. It seems like the pot was turned quite low as a result of my fiddling. I've turned it back up and have VCORE at 0.70 V, other readouts are identical to before. However, still no dice on seeing the unit mine yet.
legendary
Activity: 1736
Merit: 1006
Not underneath the heatsink, below the heatsink Slightly closer to the USB connector than the edge of the heatsink. The reset testpad is the pad on the back labeled "reset".

A visual inspection doesn't show any clearly damaged areas, but I've included more pictures just in case.  https://imgur.com/a/tOHbY

I measured the voltage using the reset testpad and these were my readings:

VCORE: 0.55 V
1V8: 1.80 V
0V9: 0.48 V
RESET: 1.61 V


im not sure but it looks like the pot is turned all the way down.. this isnt enough to run the compac..

id turn the pot clockwise until the flat part is pretty much on the bottom _ this is around .7v
epg
member
Activity: 63
Merit: 10
Not underneath the heatsink, below the heatsink Slightly closer to the USB connector than the edge of the heatsink. The reset testpad is the pad on the back labeled "reset".

A visual inspection doesn't show any clearly damaged areas, but I've included more pictures just in case.  https://imgur.com/a/tOHbY

I measured the voltage using the reset testpad and these were my readings:

VCORE: 0.55 V
1V8: 1.80 V
0V9: 0.48 V
RESET: 1.61 V
legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
Not underneath the heatsink, below the heatsink Slightly closer to the USB connector than the edge of the heatsink. The reset testpad is the pad on the back labeled "reset".
epg
member
Activity: 63
Merit: 10
Check the voltage on the reset testpad. There's a resistor and capacitor in the right corner immediately below the heatsink, make sure they're intact.

Sorry, I'm not too experienced with this. Am I looking for a resistor and capacitor underneath the heatsink after I remove it or before? Which are the reset testpad?

Pictures: https://imgur.com/a/tOHbY
legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
Oh, okay. I wasn't sure if that was a general showoff or if it was showing off for the overclock. Also it was fairly general, since a lot of the last page was just folks talking about the overclock competition.
legendary
Activity: 980
Merit: 1001
aka "whocares"
Oh hey, cool. Say didn't someone start a separate thread for this?

I love the bit of sarcasm, actually I started the thread for the overclock competition but since about 117 posts on this thread were "show off" posts I thought I would add to it.  Apparently, that is frowned upon, or at least when I post them.  


Deleted the post
legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
Oh hey, cool. Say didn't someone start a separate thread for this?
hero member
Activity: 735
Merit: 500
★YoBit.Net★ 350+ Coins Exchange & Dice
thats cool looking from what i can see being i cant see it in high rez
legendary
Activity: 980
Merit: 1001
aka "whocares"
The voltage output of that Buck is too high you need 0.6V to 0.8V maybee 0.9V if you are really pushing things.
Rich

Wanting to use DI/Alcohol for cooling and hoping to push the voltage up past 1 volt
legendary
Activity: 980
Merit: 1001
aka "whocares"
I am preparing for the big contest - I understand I am voiding the warranty Grin - Is there a way to bypass the buck converter on the compac with the device below.  My specialty is cooling not components, any help would be appreciated.  I am going to do a little extreme cooling on it and want full control of the power  Shocked

If I need to hack up the compac then that is ok

The voltage output of that Buck is too high you need 0.6V to 0.8V maybee 0.9V if you are really pushing things.

The onboard buck is very good and well overspecked for a single BM1384. Other than reducing the series resistor on the pot to allow you to get to 0.8V and adding a voltmenter no need to do anything else.


Rich

Sorry, I posted the wrong picture.  The one I was looking at is adjustable down to 0.5 volts.  If there is no way to hack the external device in I will probably reduce the series resistor.  


If you want to push more than about 17A into the ASIC (note the S5 runs maybe 11A on stock settings) you should replace the inductor. But before you worry about that you'll probably be worrying about your USB jack.

And replace the inductor and hard wire the USB port Wink
legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
If you want to push more than about 17A into the ASIC (note the S5 runs maybe 11A on stock settings) you should replace the inductor. But before you worry about that you'll probably be worrying about your USB jack.
hero member
Activity: 588
Merit: 500
I am preparing for the big contest - I understand I am voiding the warranty Grin - Is there a way to bypass the buck converter on the compac with the device below.  My specialty is cooling not components, any help would be appreciated.  I am going to do a little extreme cooling on it and want full control of the power  Shocked

If I need to hack up the compac then that is ok

The voltage output of that Buck is too high you need 0.6V to 0.8V maybee 0.9V if you are really pushing things.

The onboard buck is very good and well overspecked for a single BM1384. Other than reducing the series resistor on the pot to allow you to get to 0.8V and adding a voltmenter no need to do anything else.


Rich
legendary
Activity: 980
Merit: 1001
aka "whocares"
I am preparing for the big contest - I understand I am voiding the warranty Grin - Is there a way to bypass the buck converter on the compac with the device below.  My specialty is cooling not components, any help would be appreciated.  I am going to do a little extreme cooling on it and want full control of the power  Shocked




If I need to hack up the compac then that is ok
Pages:
Jump to: