A3000 Intermittent Heat Related Problems?

[Brosol]

My A3000 has a history of over-heating related problems which causes  system instability, random crashes & lock ups during large file  transfers & booting up.

The problems seem to be caused by overheating of the chips on the Zorro Slot side of the case.

I tried to fix this by:

- Installing 400W PSU with high CFM fan

- putting heatsinks on every heat producing chip (CIA, Amber, SDMAC,  SBuster, 68030 CPU, chips next CHIP RAM area in front of the LED, U714  gets very hot)

- installing fans, PSU fan, exhaust slot fan

- replaced GALS U202, U203, U714

Despite all this the A3000 still has SCSI R/W errors & lock ups  problems. I placed a temp probe on U714 which gets very hot.  Problems  start to show up when the temp reaches 34plus deg C.  Its a problem  during the summer when the ambient temp is too high for fans to help or  when running with the case cover on.  I not sure if U714 is the culprit  since it was replaced new or is the indicator of other problems

Now my Deneb is acting weird despite U714 temp probe showing a cool  22degC.  Its showing R/W error problems booting and copying files  between Deneb USB drives, file transfers between Deneb USB and A3000S  SCSI drives.  

I removed all cards exept the Deneb.  Swaping Deneb in ZorroIII slots  seems to help temporarily but the intermittent stability problems  persist.

Could bad aging caps be the cause?  What are the signs of bad caps?  I  wonder if A3000D motherboard caps replacement (offered by Cosmos) would  help fix the instability problems?

[Brosol]

Hi,
   A few dozen things come to mind on this: first start with the psu, check the 5v rail when your are loading it heavy,monitor it with a meter..it should maintain about 4.92v or better(some say 4.80's are ok,but my experience says otherwise).

>>>> 5v Rail: 5.03V during large file R/W

 there are many many bad psu's out there,and 400W doesn't mean anything, its more important what the rails can handle individually.I've seen 400W pc power supplies that should be rated 250w..
Cheap/bad caps in the psu can also put out dirty power and cause lotsa troubles.Check the motherboard connector and the pigtail from the psu, notorious for having a bad 5v connection.checking power before the connector on the psu and somewhere near the power connector on the motherboard should show close voltage.

>>>> seems to be OK

Assuming thats good, reseating chips-pushing on them is useless, pull them and reseat them(i doubt this is the trouble though) but we have 20 year old contacts here!Use a proper plcc puller, on those old sockets!.

>>>> Have not done this yet....need a plcc puller to prevent possible damage to chips and sockets.


The pals and stuff on the side of the board you mention run quite hot normally,but usually run along happily that way.

I don't think heat is your problem but may aggravate the situation if you have a borederline component.  

>>>> I think so as well, as the unknown problem becomes worse its becoming more heat sensitive

Bad ram may very well be the trouble,and hot ram chips can be a sign of bad ram.This could cause most of the problems you mention and have effects system wide.

>>>> The chip ram is not hot, I just put heatsinks there as a precaution since it right next to the hot chips.  Ram tests via SCRAMmer 37.3 were OK.

The wrong mask/max transfer values could also cause some of the troubles(may be a combination of things)But you said this combo ran ok before?.
Improper termination(check the diodes are the right way around,some 3000's had them soldered in wrong giving no term power to the scsi).Do you have the latest WD -08 scsi chip or the amd version?

>>>> Using -08 WD SCSI chip, mask/max transfers & termination are OK.   Problems pop up when the temp probe on U714 heatsink indicates 34-37+ Deg C temps.  I have especially have problems running it during the summer months

Do you have a battery installed and looked at the battmem?Maybe a scsi setting is off there but this is doubtfull?

>>>> Battery was replaced, no acid damage, SCSI settings OK.
It could be your buster chip is flakey,you don't mention what revision??

Usually when you replace the PAL chips with GAL's they run alot cooler.Are you sure the replacements are GALS? Are the GAL replacements at least as fast or faster than the originals?maybe a bad replacement chip?

>>>> I got GAL U202, U203, U714 replacements from....Anthony Hoffman so I'm sure its ok.

Capacitors could be going bad,especially after nearly 20 yrs, but i'm leaning toward bad ram,flakey psu or a flakey buster? Btw,what rev buster do you run?

>>>> Using Rev-09 SBuster with SB-09 flash Deneb support.  My recent Deneb instability problems don't seem to be related to overheating....U714 temp probe showed a cool 20deg C during Deneb problems.

>>>>Things to try: reseating the chips, replace SB-09 with SB-11 which is supposed to be more stable with Deneb, try another PSU, Caps replacement (what would be symptoms of aging caps?)
 

This probabaly doesn't help much,being able to substitute parts with known good helps.
If you find the trouble,please report back on it.

>>>> Its very helpful....much appreciated:lol:

Mike

A

[Brosol]

Yes that exactly how the intermittent problems are cropping up, little by little becoming more unstable as its being used.

I got back to playing with my A3000 for over year now after 10yrs of storage.  Its on frequently and slowly upgraded:

A3000 25MHZ 2mb Chip, 16mb Fast
WDC 00-08 SCSI Controller
DB25 Ext SCSI Terminator
Super Buster 9
KS 3.1 Roms
Lithium Clock Battery Holder
CyberVision 64 4MB
Indivsion ECS
Deneb
FastATA 4000 MK-V==>IDE-CF Reader
A3000 SCSI==>SCSI-IDE Bridge==>IDE-CF Reader

Drive platform mod to allow 68060 HS/Fan clearance
Cyberstorm MKII, 128MB
MC68060RC50 CPU 71E41J mask
66.66 mhz Crystal Oscillator

PS/2 Usb Mouse Adapter

Modded A3000 PSU (Original Case)
 - 400W Duro ATX PSU PCB
 - 80mm ThermalTake A128025BUT Fan
 - A3000 ATX PSU Cable

Heatsinks: DMAC, 68030, 68060
 Heatsinks for heat producing IC chips
New GAL U202, U203, U714 replacement
Lyra 1 PC Keyboard adapter
MAS-Player Pro 2

[Brosol]

Ok just noticed another problem...the video output from the A3000 flickerfixer is very jittery while the Indivision ECS video is rock stable and clear.

[Brosol]

Yes, those parts are modern equivalents to the original programmable logic.  They can run faster, draw less current and as a result will dissipate less heat.

Reseating all the socketed devices is a good suggestion, plus it doesn't cost anything.  Be careful not to bend legs when re-inserting, etc.

Ok like you say its free...I will try it

Another suggestion is to remove the case cover from the A3000 to help test the heating theory.  

I'm doing this test now running without zorro cards,  doing boot and R/W file tests on CF cards.  I will post the results with pics.

Although this stops the fan drawn air circulation, it does promote convectional air flow.  One of my A3000s tends to start having heat related problems similar to what you describe after a few hours running.  Without the case cover on it seems quite reliable though.  Does have a lot of crap installed in there, Cyberstorm mk2, Picasso IV, etc, etc.  Not surprising that air circulation is poor and heat buildup is an issue.  Never did solve the problem, the A3000 only gets infrequent use as it's the backup for the main computer (A4000T).

Wow if not even you found a solution for the poor air circulation and heat build up....then there's no hope? :eek:

I found this link from a A3000 user with related heat & poor circulation problems.  Its seems to work for him but I'm not certain I wan't to hack up and alter the look of the A3000 case to such extremes.

http://www.a1k.org/forum/showthread.php?t=16295

http://www.a1k.org/forum/attachment.php?attachmentid=17354&d=1247933590

http://www.a1k.org/forum/attachment.php?attachmentid=17353&d=1247933572

http://www.a1k.org/forum/attachment.php?attachmentid=17355&d=1247933609

I was thinking about a stealthier version of this mod by cutting a hole to fit a skinny high speed 70mmx15mm 46CFM fan next to the LEDs behind the bezel.  I don't want to drill holes or drastically change the look of the bezel.  I hope the combo of the front intake and rear slot exhaust fan will provide enough airflow to dump out the heat & counter the heat build up with Zorro cards installed.  It probally won't help during hot summer ambient temperatures.

Last suggestion is send it down here to Antarctica  I've only seen it down to -30C so far, but it'll get colder during winter.  If that doesn't keep the A3000 cool then nothing will!

Or I get a AC unit and run it perpetually but I don't want to pay the power bills.  I think a running a long dryer hose to pipe in that cool Antartic air is the best solution. :roflmao:

Quick Update:  Running with case open, no zorro cards, no fans, just heatsinks & convection cooling:

U714 Temp Probe: 42.2 deg C, 5V Rail: 5.06V
System Behavior: unstable, checksum disk block errors, lock ups during booting.  Tried different known to work CF cards but same results.

[Brosol]

This totally sounds like a bad config in your SCSI transfers. Did you check the battery mem as Mechy suggested? Here's a utility to get you into it:

http://aminet.net/package/util/sys/SetBatt1_0

A3000 SCSI config:
SCSI TimeOut Short: 128ms
LUNS Only 0
Fast Sychronous On
SCSI Queues Off

...MaxTransfer values next in HDToolBox and then your CF card adapter jumper settings are suspect too.

BTW: Are you strictly using a SCSI<>IDE<>CF Card adapter (or  SCSI<>CF adapter)?  Does the computer behave when you use a good  old fashioned SCSI hard drive?

I tried FFS, SFS using default values.  What do you suggest for CF cards?

I am using A3000SCSI<>SCSItoIDE Bridge<>IDEtoCF adapter<>CF card.  I don't have any SCSI HD anymore.

I've had mixed results using CF Cards with Amiga's. The cheapo adapters you find on ePay and elsewhere are not always the most compatible and neither are several CF card brands. After all the monkeying around I've done with various HD alternatives, I still find hard drives to be the best, most reliable solutions inside  Amiga's.

Just until recently did the problems started to crop up more and more.  The Deneb related problems are new.

Test Results #2
Temp Probe: 22.5C, case open, CPU Fan blowing over heatsink, cool intake air from outside.
Behavior: seems stable, successful booting from CF and large file transfers between CFs OK without errors.  1 CF card become corrupted from the Test#1 and had to be reformatted.

Test Results #3
Temp Probe: 22.5C, case open, CPU Fan blowing, Deneb Installed in Zorro slot 3 from the top
Behavior: successful booting from CF, Fat32 USB drive recognized & mounted.  Tried to extract a large LHA from USB to CF...failed causing a reboot.
Tried Deneb in Zorro slot 1 from top....failed to mount Fat32 drive.  Tried FFS, SFS USB drives....mounts but large file transfers working intermittently.  Again this is a new Deneb problem which happening despite cool temperatures.

[Brosol]

I'm using 4GB-Timu Premium UDMA CF, 8GB-Pixelflash 233x-Pro CF, 4GB-Diesel & 8GB-X-Mini Patriot USB drives.

Filesystems: FFS, SFS, Fat32 for PC<>Amiga

[Brosol]

Yes I'm using a R-IDSC-E/R SCSI-IDE bridge and Ebay CF adapter & Ebay CF cards

http://cgi.ebay.com/CF-Card-3-5-40-PIN-IDE-SSD-HDD-Adapter-Back-Panel-/260624996786?pt=LH_DefaultDomain_0&hash=item3cae75d9b2

How do I know & choose better CF adapters?

USB drives were purchased local NCIX:

http://ncix.com/products/?sku=39651&vpn=OCZUSBDSL4G&manufacture=OCZ%20Technology

http://ncix.com/products/index.php?sku=28946&vpn=PSF8GMUSB&manufacture=Patriot

If the chips reseating doesn't work I will swap out the chip & zip and MKII ram.

[Brosol]

I personally run 2-32GB kingston pro cards off my UWscsi on the cyberstorm ppc with Acard adapters to cf adapters wired for proper DMA. its been working 100% for 2 years+ now.

I only run SFS 1.279 on my stuff. read the full archive docs included 1.277.. it suggests max transfers for scsi and ide,and if the full dma scsi values dont work,drop down as suggested.

I'm not having much luck playing with SFS 1.279.  I set things up via WinUAE and test them on the A3000. I tried suggested max transfers and mask values but still getting boot problems:

Mask:0xFFFFFFFF
Max Transfer: 0x7FFFFFFF,  0x1FFFE, 0xFFFE

What SFS values should I try for A3000 SCSI, SCSI-IDE bridge & CF combo?

[Brosol]

Hi,
   Assuming thats good, reseating chips-pushing on them is useless, pull them and reseat them(i doubt this is the trouble though) but we have 20 year old contacts here!Use a proper plcc puller, on those old sockets!. The pals and stuff on the side of the board you mention run quite hot normally,but usually run along happily that way.

I don't think heat is your problem but may aggravate the situation if you have a borederline component.  Bad ram may very well be the trouble,and hot ram chips can be a sign of bad ram.This could cause most of the problems you mention and have effects system wide.
Mike

You were right about the ram being the culprit but so was the temperature factor.

I stripped down the A3000 to most basic running setup and starting reseating all non-PLCC chips (I don't have a puller yet).  Nothing seem to work until after reseating the zip ram....the system was stable again I was able to boot, no more R/W errors or random crashes...etc.  My problem might be unique to the A3000 since it uses Zip RAM which are prone to work themselves out of their sockets due to cycles of cooling & heating.  I found this link by Robert Davis which explains the cause and the fix.

http://www.nyx.net/~rdavis/AmigaHints2.html#ZIPRAM

The Deneb is working again with unexpected benefit which may be of interest to A3000 Deneb users.  Even though I have Buster-09 and Deneb Flash ROM upgrade to support Buster-09 I was never able to use DMA mode without resulting in a ustable sytem.  I tried using Deneb in DMA mode after the Zip RAM reseating fix and now DMA works!  The system seems stable in Deneb DMA mode & I able to large filesize R/W without stability problems.  Perhaps other weird A3000 Deneb related problems might be fixed by reseating all socketed components, especially the Zip RAM.

Thank you everyone for helping!

[Brosol]

Try adjusting the flicker fixer potentiometer and see if it clears up,these variable resistors do degrade over time.

mike

That fixed it, Thanks  

[Brosol]

So the Zip's weren't actually bad themselves... just were no longer making as good of a connection? Hmm... next time I have my A3000 apart, I'll try that.

I discovered while reinserting the 16mb of zips, it was too easy to accidently bend the legs.  I used tweezers to carefully straighten any bent legs.  Be patient during reseating & examine if all zips legs are properly inserted in their sockets.  Be static safe.