Overview Features Coding ApolloOS Performance Forum Downloads Products Order Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.
Please visit our Apollo-Discord Server for support.



All TopicsNewsPerformanceGamesDemosApolloVampireAROSWorkbenchATARIReleases
Documentation about the Vampire hardware

Vampire 600V2 Issue and Query

Leigh Russ

Posts 151
12 Sep 2017 18:51


Hi

Finally after 10months I have my V600 V2 fairly stable in one of my three Amigas. One it won't work in at all, all three are rev 1.5.

I still have a few issues and wondered if these are normal and to be expected or not? The main one being that on first boot from cold it often hangs as soon as it starts loading the workbench screen. After a reset or 2 it then loads up fine (it may hang randomly on occasion).

I have however found that once the Amiga has been on for a while and the card has warmed up it pretty much boots up every time without issue. And is loads more stable.

From past experience when something works better when warm it is often a dry joint on the PCB or under the BGA, and when it expands with the heat it then makes a good connection. What are the chances of this being the case? All the joints on the ICs and components look fine under a microscope but obviously cannot see under the FPGA.

Have already cleaned up the socket and CPU with isopropanol and sanded the socket. I'm starting to rule the socket/CPU out as being the issue as it's the same on multiple Amigas no matter how many times it's cleaned or reseated. All Amigas have been recapped. Also strange how it doesn't work on one Amiga at all despite all being the same revision.

Anyone have any similar issues or ideas?



Gunnar von Boehn
(Apollo Team Member)
Posts 6207
12 Sep 2017 20:15


Leigh Russ wrote:

Hi
 
Finally after 10months I have my V600 V2

When did you get it?
From where did you get it?

What OS do you run?
Do you use your own OS compilation or do you use a tested one like this APOLLO-OS?

Do you get Lock-ups or GURUs?



Steve Ferrell

Posts 424
12 Sep 2017 21:05


Leigh Russ wrote:

  Hi
 
  Finally after 10months I have my V600 V2 fairly stable in one of my three Amigas. One it won't work in at all, all three are rev 1.5.
 
  I still have a few issues and wondered if these are normal and to be expected or not? The main one being that on first boot from cold it often hangs as soon as it starts loading the workbench screen. After a reset or 2 it then loads up fine (it may hang randomly on occasion).
 
  I have however found that once the Amiga has been on for a while and the card has warmed up it pretty much boots up every time without issue. And is loads more stable.
 
  From past experience when something works better when warm it is often a dry joint on the PCB or under the BGA, and when it expands with the heat it then makes a good connection. What are the chances of this being the case? All the joints on the ICs and components look fine under a microscope but obviously cannot see under the FPGA.
 
  Have already cleaned up the socket and CPU with isopropanol and sanded the socket. I'm starting to rule the socket/CPU out as being the issue as it's the same on multiple Amigas no matter how many times it's cleaned or reseated. All Amigas have been recapped. Also strange how it doesn't work on one Amiga at all despite all being the same revision.
 
  Anyone have any similar issues or ideas?
 
 

 
Very unlikely that one of the Vampire's BGA chips is at fault so soon or any traces either. More than likely the fault(s) lay with your 20+ year old Amiga.  Sounds like a classic case of bad capacitors or cracked solder joints in your A600.  Has it been re-capped?

This is one of the reasons why I'm waiting for the standalone Vampire.  When combining a modern FPGA with 25 year old hardware these scenarios are not only to be expected, but they will increase in frequency. 
 


Szyk Cech

Posts 191
12 Sep 2017 21:07


Steve Ferrell wrote:

Has it been re-capped?

What the stupid question! He wrote she is re-caped!


Steve Ferrell

Posts 424
12 Sep 2017 21:11


Szyk Cech wrote:

 
Steve Ferrell wrote:

    Has it been re-capped?
   

    What the stupid question! He wrote she is re-caped!
 

 
My what a helpful reply!  Thanks for your insightful comment.  I'm sure Leigh will appreciate your remarks as much as I did.  I missed the part where he stated it was re-capped, but again, the rest of my statement stands on its own merit....unlike your moronic response.
 
And just because it has been re-capped doesn't mean that the job was done well or that some of the replaced caps haven't again gone bad.


Leigh Russ

Posts 151
12 Sep 2017 21:37


Gunnar von Boehn wrote:

  When did you get it?
  From where did you get it?
 
  What OS do you run?
  Do you use your own OS compilation or do you use a tested one like this APOLLO-OS?
 
  Do you get Lock-ups or GURUs?
 

Got it in December from Kipper2k after waiting for ages on the waiting list.

Running OS3.9 BB2, have tested both with ClassicWB and without. FBlit, FText, Cyberbugfix, etc are disabled.

Same issue occurs both on GOLD2 and Gold2.5

Haven't tried ApolloOS on it as after finding there's errors in the image, I mounted it directly in WinUAE to try it instead and personally did not like it at all. Found it pretty bloated and overkill so much prefer to build my own to my taste and needs

It doesn't guru, just freezes and locks up.

Steve Ferrell wrote:

And just because it has been re-capped doesn't mean that the job was done well or that some of the replaced caps haven't again gone bad.

Yep, all three Amigas have been recapped with Panasonic electrolytics (none of the ceramic or tantalum stuff here) last month. All identical values and voltages as the originals as per the original bill of materials.

Unlikely to be capacitor related when it's the same across all three A600s though. That would be extremely unlucky.

Same issue on all the amigas, well apart from the one it doesn't boot on at all (maybe like 1 in 10 you may get it to boot if your lucky). The only difference I see with that one to the others is slightly different numbers on the Gayle chip.

CPU has also been reflowed on one just to make sure too.

As for it unlikely to be the BGA. It's actually a more common issue than you would think After all, look at the red ring of death on the original XBox 360 issue. If you overheat one of those so it's nice and toasty (the wrapping in a towl trick) then they often worked fine. The issue with them was exactly that, poor soldering/solder used on the GPU :(. It was just a thought anyway after finding once it's been on half an hour and running hot it's proper stable

If need be I guess I can live with it but the restarting several time to get it to boot up the first time can get annoying.

Oh, the KS ROM has also been removed as leaving it in just causes more issues




Steve Ferrell

Posts 424
12 Sep 2017 21:54



As for it unlikely to be the BGA. It's actually a more common issue than you would think After all, look at the red ring of death on the original XBox 360 issue. If you overheat one of those so it's nice and toasty (the wrapping in a towl trick) then they often worked fine. The issue with them was exactly that, poor soldering/solder used on the GPU :(. It was just a thought anyway after finding once it's been on half an hour and running hot it's proper stable

Oh man, please don't compare Microsoft quality control with the Vampire team's!  That would be insulting to the Vampire team! LOL!

As you say you're having the same problem across all 3 Amigas it would lead one to suspect the Vampire board.  I wonder if the CPU connection between the Amiga and the Vampire is the culprit?  There could be a problem with the socket on the Vampire being too loose until things warm up, but from what I've heard, the Vampire socket is a really tight fit in most cases.

It might also be worth taking a pencil eraser and cleaning any oxidation from the 68000 CPU legs since the socket clamps over them. 


Roy Gillotti

Posts 517
12 Sep 2017 22:10


What are you using for a power supply? A stock A600/1200 PSU is likely to not be able to keep up to the additional power consumption. I had stability issues when using an 80&90Watt PicoPSU as well. My current stable setup is using an industrial PSU, a Mean Well T60-b, some I've talked to on IRC use a different, but comparable model.


Leigh Russ

Posts 151
12 Sep 2017 22:40


Roy Gillotti wrote:

What are you using for a power supply? A stock A600/1200 PSU is likely to not be able to keep up to the additional power consumption. I had stability issues when using an 80&90Watt PicoPSU as well. My current stable setup is using an industrial PSU, a Mean Well T60-b, some I've talked to on IRC use a different, but comparable model.

I'm currently using my A500 PSU (the heavier slightly more powerful one).

I also got a PicoPSU. I designed an Amiga style case for it https://www.thingiverse.com/thing:1794271 and use it with Ian Stedman's ATX adapter. Unfortunately my 12v PSU is no more till I get another one.

I got a quote from Meanwell for on of their PSUs but their postage was as much as the PSU so was contemplating one of these...

EXTERNAL LINK 
Quite a bit cheaoer but don't know if they any good.



posts 9