Hello all, I hope someone is able to help me get this working!

I've recently bought an AT-500 Megabody off of eBay for my A500 (Rev 6A), but I cannot get it to work at all. Here's a generic photo of the card taken from amiga.resource.cx:



What happens is this:

I power the Amiga on, and it all looks pretty normal regarding screen colours...

black --> grey --> very light grey (which is almost white) --> white.

The problem is it never gets past white, it just sits there and flashes the Amiga power light once (led filter goes off then on), and the floppy drive never clicks once either (the thing it does when the tick / floppy animation appears). From this fault state I am able to reboot the Amiga and the same fault will repeat as before. This is with a Kickstart 2 ROM chip (37.299).

I first tried the Megabody though with a 1.3 ROM chip (intended use), and this was a few days ago now so I can't remember precisely what occurs but it was basically the same sequence of screen colours and it sticks on white but the Amiga power light (led filter) then continually flashes instead of just the one flash and then I think that it reboots and the cycle then repeats (it may even turn to a black screen as the power light flashes but I can't remember). Anyway, the reason I then tried the 2.0 Kickstart was that I was hoping I might be able to get into the Amiga Early Startup Control (well KS2 equivalent) but nope, it doesn't get that far which makes sense as there is definitely a problem preventing it getting to that stage.

There's 4 jumpers/connections on the card:

--MEM
--DTK
--LED
--Kick 1.2

I have MEM and DTK jumpered as per manual.

When I connect an LED to the LED connection the behaviour changes slightly - instead of staying on a white screen with one power light flash, what I get is 5 flashes followed by one long flash and then the screen turns very black (no video?) and it stays on the black screen but keeps repeating the power light flashes (5 short flashes followed by one long one). I'd call that 6 flashes, but some people may see that differently to me. The actual har drive LED never lights.

When I jumper the Kick 1.2 the colour sequence goes the same but instead of ending on white I get a light lilac/pink colour instead.

It makes no difference whether I have my hard drive connected or not (Hitachi 1.8GB 2.5" DK213A-18). The hard drive does spin up as expected when it is connected.

With the Megabody removed, I have checked the continuity between the Megabody CPU pins (the ones that make contact with the Amiga CPU socket) and the CPU pins of the 68000 when plugged into the socket on top of the Megabody. They all measure good except for two pins - PIN 6 (D0) and PIN 10 (DTACK). I expect this is a design feature of the card as those pins appear to go off to other parts of the Megabody circuit. If anyone knows this not to be the case then please correct me and I'll have a closer look. IIRC, Pin 6 goes off to other circuit parts and Pin 10 either goes somewhere internal or nowhere at all (from looking at the board).

I have tried two seperate PSU's, one of which provides plenty of power as it's a 200W PC PSU. Again, it makes no difference.

On close inspection of the card, there are some marks. Right below the where the CPU fits is a couple of marks, one of which has exposed some copper on the track - but I have checked continuity and it's okay. Also, those SMD resistors at the bottom have taken a bit of a beating too as I see little dents in the solder at the top of those resistors where a screwdriver shaft has been pressing against them (to lever out the CPU) - but again, it's purely cosmetic I believe and I have checked continuity and resistances as best I can and they all check out okay.
There is a 470 Ohm resistor just below the jumpers - this is the only 470 Ohm resistor on the board but it does look different as in the smooth surface at the top looks a bit bubbly and the writing is blurry which made me think perhaps it's overheated and it's resistance has shrunk, but no...it measures exactly 470 Ohms. I'm assuming this is for the LED.

The only thing I haven't checked yet is the continuity between the Amiga CPU socket and the Megabody pins when plugged into the Amiga, but I'm confident they're good as the fault behaviour hasn't changed and the card has been attached and removed a few times now.

Here's the auction URL so that you can see the actual photos of the card in question: http://www.ebay.co.uk/itm/282327866079

Information on the card: http://amiga.resource.cx/exp/at500

User Manual direct download (German): http://http://amiga.resource.cx/manual/AT500_Megabody.pdf

If you look at the actual photos of the card I purchased, you'll see a green LED wire incorrectly connected to the top Kick 1.2 jumper pin - Of course though, I corrected that before hooking it up and powering on.

Thanks in advance.

Thanks for your reply Darkage. It should work as it was sold as 'used', and with Google Translator I saw no reference to it not working. I have contacted the seller and he said it was working with a Kickstart 1.3 A500 with a hard disk that he doesn't have any more, and that's all I know about it.

When it didn't work I did clean it and thoroughly inspect it all. Some of the solder joints on the CPU socket underneath look shinier than others, and there are tiny bits of solder around the pins in this area along the card which are very small admittedly, but could indicate previous touch ups - or, it may have been hand soldered that way - I don't suppose these were mass manufactured so that would make sense. Plenty of flux residue too but it doesn't mean anything. It all looks good, but either way the Amiga is having none of it.

I measured the voltages on the visible chips and they are all good @ 4.94V. Is it possible the GALS have lost their programming? According to the spec sheet they are only guaranteed for 20 years between programming, and the card is 22 years old.
The socketed ROM chip's programming is guaranteed to last 200 years however. So once those GAL'S have gone they're gone unless someone has kept the code somewhere? Very sad.

I could heat joints but it's too invasive as I may have to return it for a refund. Tomorrow I'll probably take the motherboard out of it's shield and check the continuity from bottom side to Megabody CPU.

P.S. I watched the UV video and that was quite fascinating to me. Especially the part where the UV exposed the heat coming from the high amps (shorted) trace. This must have something to do with the infrared that camera's can detect....unless of course the trace was giving off UV as a result of the short - which just doesn't sound right does it? Interesting how the heat spread from one direction to the other...very very interesting.

yes the GALs have a limited lifetime.   there are some hacks / brute force methods that you can use to read them, but of course not guaranteed.    other than that would need to do reverse engineering with logic analyser and fingers crossed.

Maybe check what revision of motherboard seller was using ?  wondering if its a rev 6 as well ?  is your stock standard or existing memory expansions present ?

Any other A500 users nearby, I would definitely be tempted to try in a different machine with different CPU as well.

Silly question, your machine is confirmed working without the AT500 plugged in ?

With the UV light its all about light wavelengths.  some objects reflect different things not normally present in visible light, plus camera lens is a little bit more sensitive to UV light and picks up things that the eye doesnt see.      Ive got a 030 accelerator for my a500 that doesnt work, cant wait to do UV inspection

If you do end up returning it, why not just go for one of these modern boards ?  http://www.kipper2k.com/a500fastmem.html

I only have  a 512KB Chip RAM in the trapdoor, nothing else. I did try it without that and it still gets stuck on the white screen.
Revision 6A is the most common in Europe I believe, but even if it didn't work with a 6A I only have the one A500 which leaves me pretty stuck. I wish my A500 didn't work without the AT-500...but alas, it works great without it. I can't believe I just said that.

I checked the continuity between the motherboard CPU socket and Megabody card CPU socket (which was a pain in the back side btw) but all is good there as I expected (except for pins 6 and 10 as previously explained).

Kipper's card looks good, but didn't he throw in the towel? Last I heard he was finishing off the Vampires and then no more. I hope I'm wrong though.

Just an update here, as I managed to fix this last week. I had another look in the area where there were marks on the board and damage from someone removing the CPU with what I guess was a screwdriver and no cloth or protection to protect the pcb. There was a cracked solder joint I missed on a leg of IC9 (hidden under CPU after my first inspection). IC9 must have been used as a lever point to prise out the CPU (damage visible on IC9 proves this, and the IC slants down to the right where the cracked joint was). Once I repaired that, the machine came ready, but still I couldn't get it to see my hard drive, nor was the mtec-at500.device appearing in memory.

Over at EAB, Toni Willen was so kind as to check the ROM images he had me make and he confirmed all was well with the ROM. He examined the ROM and said that the mtec-at500.device will not appear unless a hard drive is detected. This made me take a look at my IDE cable - so I replaced it and hey presto...the hard drive booted! I can now move ahead with my A500 system.

Thanks Mike. Did I tell you this is remaining a Kickstart 1.3 machine?