Cleaned floppy drive heads and cleaned/relubed gliders:
Format floppy to get the drive exercising:
Remove wait states to increase system performance now we have a fast ROM:
Do a quick benchmark:
The result is dragged down a bit by the old MC68881FN16B FPU but otherwise the entire system is a little bit quicker.
REMINDER - Stay logged in for at least 2 hours a week to get whitelisted.
Also it helps build a picture where our "good traffic" is coming from for detection scripts.
:o)
Also it helps build a picture where our "good traffic" is coming from for detection scripts.
:o)
Falcon B52803000012 service blog
-
Steve
- Moderator

- Posts: 3334
- Joined: 15 Sep 2017 11:49
Re: Falcon B52803000012 service blog
You do not have the required permissions to view the files attached to this post.
-
Steve
- Moderator

- Posts: 3334
- Joined: 15 Sep 2017 11:49
Re: Falcon B52803000012 service blog
Damn this was a very tough Dallas clock to remove. Every pin was fully desoldered and loose but it wouldn't budge. So I protected the area with kapton tape and used my hot air gun and after a little while it gently fell out :)
You do not have the required permissions to view the files attached to this post.
-
stween
- Site sponsor

- Posts: 282
- Joined: 08 Sep 2018 15:10
- Location: Brooklyn & Edinburgh
Re: Falcon B52803000012 service blog
You have the same RAM access performance as I have :) (I haven't gotten back to looking at this yet!)Steve wrote: 25 Apr 2021 19:27Do a quick benchmark:
PXL_20210425_180517232.MP.jpg
The result is dragged down a bit by the old MC68881FN16B FPU but otherwise the entire system is a little bit quicker.
-
Steve
- Moderator

- Posts: 3334
- Joined: 15 Sep 2017 11:49
Re: Falcon B52803000012 service blog
Okay here we go. All hardware work is done apart from installing a CF card. Just waiting on his backed up Atari image to arrive. But other than that, she is absolutely pristine and singing like a bird!
Fitting new NVRAM:
Building & installing Exxos 4700uf reservoir cap:
Board fully recapped, left mid right:
Clock patch installation:
New fan wired to 5v for silent operation:
Overview of finished falcon:
Case and keyboard fully cleaned:
Fitting new NVRAM:
Building & installing Exxos 4700uf reservoir cap:
Board fully recapped, left mid right:
Clock patch installation:
New fan wired to 5v for silent operation:
Overview of finished falcon:
Case and keyboard fully cleaned:
You do not have the required permissions to view the files attached to this post.
-
photonstorm
- Posts: 4
- Joined: 25 Mar 2021 23:40
Re: Falcon B52803000012 service blog
Looking beautiful :dualthumbup: - can't wait to get this back and start playing with it!
-
exxos
- Site Admin

- Posts: 28619
- Joined: 16 Aug 2017 23:19
- Location: UK
Re: Falcon B52803000012 service blog
:bravo: :2k2:
-
centuriontech
Re: Falcon B52803000012 service blog
While I am not a big fan of radial caps in axial caps designed application (for many reasons), I must admit the job is nice and tidy.
Just a side note about your fan installation:
There is a _very_ good reason, why fans are not connected on main power rail with rest of the system, by design. They are rather powered from either entirely dedicated power supply, or at least from secondary/utility power rail - just like in Falcon. This design rule is golden especially for digital processing applications, RF applications and audio devices. So pretty much everywhere ;) The reason is motors induce a lot of electric noise back into the power rail, resulting in whole bunch of issues. If the system design limitations otherwise enforce DC motor on main power rail, it is always connected through properly calculated low-pass filter (L-C, C-L-C,...). Still, it is always suboptimal, because no filter can filter out entire frequency spectrum in all motor operating conditions.
Just a side note about your fan installation:
There is a _very_ good reason, why fans are not connected on main power rail with rest of the system, by design. They are rather powered from either entirely dedicated power supply, or at least from secondary/utility power rail - just like in Falcon. This design rule is golden especially for digital processing applications, RF applications and audio devices. So pretty much everywhere ;) The reason is motors induce a lot of electric noise back into the power rail, resulting in whole bunch of issues. If the system design limitations otherwise enforce DC motor on main power rail, it is always connected through properly calculated low-pass filter (L-C, C-L-C,...). Still, it is always suboptimal, because no filter can filter out entire frequency spectrum in all motor operating conditions.
-
Steve
- Moderator

- Posts: 3334
- Joined: 15 Sep 2017 11:49
Re: Falcon B52803000012 service blog
@centuriontech thanks Pavel, so it would be best for me to connect it back to its dedicated 12v and put a resistor in-line to reduce its speed. I'll do that, thanks I appreciate the advice!
Who is online
Users browsing this forum: Acsi, ClaudeBot and 68 guests