Its a sad day - My upgraded BPPC wont boot :-(

[fitzsteve]

I've tried it in both A1200's and it starts to boot, a few reads of the HDD and then the power light flashes and it stops.

I can't even boot from a WB Floppy

The system's are booting fine with my other BPPC so I don't know what to check.

I've tried with and without Ram and tried different Ram.

I've tried a few different Oscilators on the 060 and that didn't help.

I'll try cleaning the contacts, I don't see how that could be the problem though as it was working fine yesterday and I haven't changed anything

Steve.

[FOL]

Originally Posted by fitzsteve

I've tried it in both A1200's and it starts to boot, a few reads of the HDD and then the power light flashes and it stops.

I can't even boot from a WB Floppy

The system's are booting fine with my other BPPC so I don't know what to check.

I've tried with and without Ram and tried different Ram.

I've tried a few different Oscilators on the 060 and that didn't help.

I'll try cleaning the contacts, I don't see how that could be the problem though as it was working fine yesterday and I haven't changed anything

Steve.

PSU?

[fitzsteve]

Originally Posted by FOL

PSU?

I have two BPPC's and the other is working fine in both A1200's

The overclocked one will not work in either so I don't think its power related.

I've been using it almost daily until today without any issues.

[delshay]

Originally Posted by fitzsteve

I've tried it in both A1200's and it starts to boot, a few reads of the HDD and then the power light flashes and it stops.

I can't even boot from a WB Floppy

The system's are booting fine with my other BPPC so I don't know what to check.

I've tried with and without Ram and tried different Ram.

I've tried a few different Oscilators on the 060 and that didn't help.

I'll try cleaning the contacts, I don't see how that could be the problem though as it was working fine yesterday and I haven't changed anything

Steve.

now lets see,oh don't tell me,it'`s the one overclocked at 330Mhz am i correct?

[Retrofan]

Yes, yes, it's a sad day... you break your BPPC, I lose my pencil...
(excuse me, bad sense of humor, I mean that's a great loss).

Have you asked Stachu?

[Vairn]

poor PPC.

but remember any piece of hardware which runs outside its normal specs, will fail, be it a day, or 2 years after it was done, for it to happen on a O/C PPC is sad, but to "quote" a famous man. Most hardware gets alot more testing then these PPC have had.
"Cap'in The Engines canna take the strain"

[Echo]

I push my Core i7 to its limit, but overclocking rare legacy hardware is surely madness?

[Damion]

@fitzsteve

One idea - are you supplying juice to the floppy connector? I imagine you are, but mention it only because it's one reason why the upgraded card may fail when the stock one does not - it's drawing more. Regardless, fingers crossed it's something silly and you get her up and running soon.

I push my Core i7 to its limit, but overclocking rare legacy hardware is surely madness?

Generally I agree.

A matter of overclocking only the CPU (like the popular Apollo 1260/80MHz setup) is one thing, but frying logic with memory/SCSI overclocking means the card is basically a paperweight.

I can't say I know how the Blizzard/Cyberstorm PPC cards work in this regard. On the Cyberstorm at least, the memory bandwidth seems to increase significantly when overclocking the 68k CPU, even when using the 3rd oscillator. This seems to indicate overclocked RAM, which is a huge gamble IMHO. I have no idea how this plays out when overclocking the PPC, but would be very interested to hear the details if anyone knows.

One suggestion is to measure the temperature of the FPGA's (and other chips) before and after the overclock. If there's any significant increase, it might be wise to at least cool them. I can't remember the specifics, but on my TekMagic the increase was enough for me to leave it stock. Checking the datasheets for the logic on your particular card (and utilizing a scope) is another good idea IMHO. Another thought - I'm no expert, but surely the logic is programmed with certain timings in mind for a particular clock rate. So many variables.

I may be getting old, but WRT old hardware, I'm finding that leaving things well alone and sleeping easy at night is the better alternative. :P

[delshay]

Originally Posted by Damion

@fitzsteve

One idea - are you supplying juice to the floppy connector? I imagine you are, but mention it only because it's one reason why the upgraded card may fail when the stock one does not - it's drawing more. Regardless, fingers crossed it's something silly and you get her up and running soon.

Generally I agree.

A matter of overclocking only the CPU (like the popular Apollo 1260/80MHz setup) is one thing, but frying logic with memory/SCSI overclocking means the card is basically a paperweight.

I can't say I know how the Blizzard/Cyberstorm PPC cards work in this regard. On the Cyberstorm at least, the memory bandwidth seems to increase significantly when overclocking the 68k CPU, even when using the 3rd oscillator. This seems to indicate overclocked RAM, which is a huge gamble IMHO. I have no idea how this plays out when overclocking the PPC, but would be very interested to hear the details if anyone knows.

One suggestion is to measure the temperature of the FPGA's (and other chips) before and after the overclock. If there's any significant increase, it might be wise to at least cool them. I can't remember the specifics, but on my TekMagic the increase was enough for me to leave it stock. Checking the datasheets for the logic on your particular card (and utilizing a scope) is another good idea IMHO. Another thought - I'm no expert, but surely the logic is programmed with certain timings in mind for a particular clock rate. So many variables.

I may be getting old, but WRT old hardware, I'm finding that leaving things well alone and sleeping easy at night is the better alternative. :P

the card should be ok,but is hard to tell from here. there maybe timing issues which i have detected here and fixed on my set-up,but this is not the fault.

all modification on the card will have to be checked but im 90% sure i know what wrong with it. if im right it's nothing you can do. you will have to send it out to repair.

the card is repairable *if* there is a hardware fault.

[Zetr0]

@Damion

There are no FPGA's (field programmable gate array) on the BPPC, just CPLD's (complex programmable logic device).



@thread

Reviewing the components on the BPPC they are within tolerance - the PPC chip is designed to go faster that 330Mhz - its clock is isolated from the rest of the board, so to be honest I do not believe that this is the fault anyway.

I suspect that something discrete (and old) has decided that 13 years was enough. (which by today's cheap throwaway electronics is amazing)

Looking at the design of the BPPC I suspect it could be something to do with internal power regulation - the up/down inherent nature of the problem leads me to believe that (internally) there is a power distribution fault.

At this point (if this was my card) I would look at replacing the two regulators and the discrete (805) tantalum capacitors under the PPC heat-sink - of which there is 14 of them.

From here if there was still issues with testing I would look at changing the 4x 100uf 10v 1206 capacitors

(this is all on the underside)

Eventually I would replace all the discrete components - there's not that many to be honest.


However, Stan would be the best to ask for information on these adapters as he has infinitely more experience than I probably ever will have

[fitzsteve]

Thanks for all the info and support guys

I'm pretty sure we can rule out Power Supply as the issue. As the my other BPPC + a BVision are running fine in both setups.

Now there is some more info after speaking by email with Stan.

He asked me to check a few things and without any IDE connected the card boots immediately to a KS screen but does not attempt to boot not does it see any SCSI if I attach my SCSI HDD (which again is working on the other BPPC) it does not even see the PCCSCSI.device in the boot menu.

Annoyingly I cannot enter the BPPC Menu with ESC, it just wont work and I tried this on my Desktop A1200 as well in case it was the PC Keyboard being a pain but no joy there either.

Now when I connect an IDE device or try to boot from floppy it try's to boot as I mentioned before but after initial startup, i.e the 3.9 rom reboot it gets stuck in infinate black screen reboot or with 3.1 just hangs at the black screen.

If I try to boot WB3.1 Install from floppy it GURU's with 8000 000B I think this is CPU/FPU related?

I've sent some extra info to Stan so just waiting now to hear back with my fingers crossed

Steve.

[TCD]

Originally Posted by fitzsteve

If I try to boot WB3.1 Install from floppy it GURU's with 8000 000B I think this is CPU/FPU related?

Yep :

If the first digit of the subsystem ID is greater than 3, there is no way to recover from the error. In these cases, subtract 8 from the first digit to get the subsystem ID number.
Subsystem ID codes
CPU Trap*00

http://www.amigahistory.co.uk/guruguide.html

[fitzsteve]

Stangely I just loaded up a floppy based game ok, Silkworm was working!

This is with BPPC and no Ram, nothing else in A1200

Going to try basic workbench 3.1 with the 060 libs, brb.

Edit, tried that and it Guru's the same way as the WB Floppy.

[Vairn]

could it be the 060 that is dying? not the BPPC?

[Damion]

Originally Posted by Zetr0

@Damion

There are no FPGA's (field programmable gate array) on the BPPC, just CPLD's (complex programmable logic device).

Thanks for the correction, looking at the board I can see that. (I've never owned one). Anyhow, my point was that overclocking the CPU often goes tandem with overclocked logic, which is obviously a questionable practice.

While the PPC appears to have a separate oscillator effecting only itself, that's not the case with the 68k, which also overclocks the RAM. I didn't intend to barge in and blame the failure on overclocking, I was more so responding to Echo above. Anyway, stressing aging components beyond their design parameters is a factor to consider.

[fitzsteve]

Originally Posted by Vairn

could it be the 060 that is dying? not the BPPC?

I guess its possible, I've been running it mostly at 66mhz but sometimes at 72mhz and didn't use a fan, so perhaps I cooked it.

Its latest mask though and it wasn't hot to touch, cooler then my 50mhz XC 060 on the other BPPC anyway!

Annoyingly its soldered to the board so no easy switcheroo

[DDNI]

oh noes! I definitely felt a great disturbance in the Force, as if millions of circuits cried out in terror and were suddenly silenced... I really feel for you.


I truly hope you get it fixed.
Sadly I do not have any advice to offer.

It should be worth pointing out that the 330mhz PPC is not the OLD original PPC but a new one rated at 300mhz. Therefore the overclock that some are discussing is not as great as you may think.

[fitzsteve]

Hey thanks for the support DDNI.

Stachu thinks Flash IC is faulty and has sent me instructions to reflash, if that fails there is a 2nd option (replace the IC with new flashed unit)

So hopefully it will live again soon!

I need to ask a favour (if you're reading Prowler) Can I borrow a fully 100% working and reliable external FDD?

Cheers!

Steve.

[delshay]

is this set-up in a desktop A1200 or a tower.

[fitzsteve]

Originally Posted by delshay

is this set-up in a desktop A1200 or a tower.

This was in the Tower, but I also tried the card in my Desktop which also has a BPPC/BVision.