Can I buy ACE2B Without an Agnus ?

ljmarent · Jun 13th 2022

Hi Jens

I've built up an Amiga 2000 with ZZ9000 RTG and sadly the ZZ doesn't handle NTSC modes very well, some not at all..
Anyway ...

My quick fix is to switch over to a 2MB PAL Agnus

As luck would have it, I have two of these 318069-10 Agnus chips on hand...
I also have a few new ACE2b's on hand... but I'd rather not remove the pretty sticker and have leftover NTSC 8375 chips laying around.

So... what I'd like to do is buy an ACE2b with C9 populated and no Agnus chip in the socket.

Can we do something like that ?

Jens · Jun 14th 2022

We need to test the unit, then remove the Agnus and still provide warranty. As I explained before in this forum, this sequence will result in higher cost than you just buying the NTSC version in the shop.

ljmarent · Jun 16th 2022

Okay,

No problem,

I had searched briefly for another mention of this, but not found it, so hopefully it will be easier for the next guy to find two such questions.

I've modded one of my ACE2b's

Added the C9 0603 100nF cap

NOTE to others:
Using a Hakko PLCC-84 extractor tool, the MOS Agnus chip and fairly new socket required an 'unusual' level of force to extract.
I've extracted dozens of PLCC-84 chips using this socket specific tool and this was by far harder than extracting chips that have been in sockets for 30 years. Just be aware if you are going to do this yourself, the Agnus chip and iComp selected socket seem to form more than a mechanical bond.

NOTE to iComp:

The ACE2b I modded was hT6Ua for your records aka. cancelling warrantee

Jens · Jun 17th 2022

Quote from ljmarent

I've extracted dozens of PLCC-84 chips using this socket specific tool and this was by far harder than extracting chips that have been in sockets for 30 years.

Thanks for confirming the quality of parts we've chosen. I know there's a lot of bad parts out there - you know what they say: Chinese manufacturers can always make it cheaper - and worse. The trick is to convince them that you want quality and you're willing to pay for it.

Our buyer lives in China, visits factories for us and knows how to identify parts that are not even worth shipping as a sample. Further, we make sure that every single person in the chain makes their fair cut - also known as "fair trade". As a result, this ongoing quest for "the better part" has lead to manufacturers actually *listening* to my suggestions for improvement.

Quote from ljmarent

Added the C9 0603 100nF cap

Note that this is ONLY required for the VBB-marked Agnus chips.

ljmarent · Jun 17th 2022

Quote from Jens

Note that this is ONLY required for the VBB-marked Agnus chips.

Yup, my NOS 318069-10 are VBB parts, or I wouldn't have.

Your choice of sockets will definitely hold their electrical contact.

For my new boards, I've been using Assman A-CCS 084-G-T gold plated PLCC sockets.
Another nice German company.

With a little luck they will hold up as well as the sockets you have sourced from the east.

It's a fetish...I love to see all those gold pins staring at me..

Jens · Jun 17th 2022

Quote from ljmarent

It's a fetish...I love to see all those gold pins staring at me..

You might want to re-think that choice. Remember electro-chemistry lessons: Gold is a precious metal, and the tin/lead alloy isn't. That means that the gold will act as a catalyst on the tin/lead alloy, accelerating oxidizing. In the long run, it's better to have contact metals "as similar as possible".

I never understood why Assmann is offering these sockets. There is hardly any production IC that has gold pins. If you want to make this work reliably, you need to gold-plate the IC pins as well. Possible, and maybe even a good choice if you live near the coast with lots of salt in the air, but overkill if you ask me. This is not a high-reliability application, and if it was, there are plenty of other things to fix first

ljmarent · Jun 17th 2022

Quote from Jens

You might want to re-think that choice. Remember electro-chemistry lessons: Gold is a precious metal, and the tin/lead alloy isn't. That means that the gold will act as a catalyst on the tin/lead alloy, accelerating oxidizing. In the long run, it's better to have contact metals "as similar as possible".

I seat my chips with Stabilant 22a, so I don't have to worry about oxidation\corrosion.
While still enjoying that lovely gold look

The chips come out a lot easier, even years later, or when exposed to salt water.
PLCC sockets are easy to replace, but damaging a MOS chip on difficult extraction is pretty well a tragedy now.

anyway... I appear to be up and running...

Jens · Jun 18th 2022

Quote from ljmarent

Stabilant 22a,

"activated by current" (probably heat - didn't bother to find the activation temperature) - and it contains some grease, too. So it ticks all the boxes of things that I generally recommend to avoid.

As you confirmed, the grease is still on years, if not decades later. Current? Not much of that on a logic pin - this is in the microamps area on most pins, with the exception of power pins.

The "22a" variant is already mixed with IPA: 5ml of "22" with 25ml of IPA makes 30ml of "22a". In essence, you clean the contacts with IPA (my recommended method, and a lot cheaper), and you pay a lot extra for things that don't do much good.

While this may work for you, and identifying+fixing an unreliable contact will surely be in your field of expertise, it's certainly not recommended to make a bad decision in the first place (precious metal on non-precious metal), only to counteract it with another not-so-good decision. This almost feels like the reasoning for prescribing combinations of hard medication, where a number of pills is meant to counteract the side-effects of another. Don't get me started on why I stay away from doctors!

ljmarent Jun 13th 2022