810 Weird 810 issue

[jaek_3]

They can be replaced with two separate caps if physically possible.

How do you mean?

[Fox-1 / mnx]

Usually, three-legged caps are just 2-in1's. Example for a 12V symetric PS:

 

o---[]|---o---[]|----o

+ - + -

 

From left-to-right:

o +12VDC __o 0V __o -12VDC

 

Using two separate 4700uF ones with 2 of the legs connected, acting as 1 leg, you'll get the same result. Double-check the schematics if this is true for your caps. However it's the most logical use of the 3 legs there may be exceptions.

 

[jaek_3]

I combine the negative leads, yes?

[Fox-1 / mnx]

No. The positive from one cap and the negative from the other cap must be tied together. That point is your 0 Volt.

[+kheller2]

I know the 288 RPM vs 300 RPM thing has been beaten to death.. but... I was just reading through the "Atari 810 Disk Peripheral Device Description" document by Harry Stewart and in that document it states the RPM was to be 300. In fact there is a section for MPI mechs and states the speed should be 300 RPM. Earlier in the document it was written as 305 RPM, then crossed out to 300 RPM.

 

Interesting history but still no definitive reason why it was dropped to 288.

 

810 FSM Version 1 has it set to 290

810 FSM Version 2 has it set to 288

[+Nezgar]

When Kevin posted that document, I also noticed that same documented RPM info: http://atariage.com/forums/topic/269694-improved-sector-layout-cx8111/?p=4011363

 

Phaeron write some pretty detailed info, that the the 810 would only have a 1% margin to fit the track layout in 1 revolution at 300RPM, instead of a 4% margin at 288 RPM.

 

ijor also made a similar rationale in a recent post, to erase the tracks with a considerable overlap "to be able to format 18 sectors comfortably without a hard index hole pulse" - http://atariage.com/forums/topic/183109-xf551-compatibility-question/?p=4181232

 

So yeah, increased reliability seems to be the main working theory, especially with the really bad power boards in the first batch of 810's which would reportedly cause big RPM fluctuations.

 

Interesting seeing the documented decline of the recommended RPM. 305... no wait, 300... try 290... and for good measure, 288...

 

Edit: and then soon enough people found a nice stable 288RPM drive had enough room to fit a 19th (phantom/duplicate) sector for copy protection, or more pushing it down even more...

 

Edit 2: adding link from Analog issue 12 page 97 about topic of poor speed regulation in pre-1982 drives: https://archive.org/stream/analog-computing-magazine-12#page/n95/mode/2up

[jaek_3]

So I know it's been quite some time since the last post in this thread, but I recently dug out my 810 only to find it doesn't work anymore! Doesn't read or write disks. Can't say I'm really surprised...

 

Anyways, the head will initialize upon powerup and sometimes when it receives commands but apart from that it sits still as a stone. 

[svenski]

It may just need a clean and lube of the drive rails. Check if the head mechanism is moving.

 

l’be had numerous drives where a quick clean has returned them to service.

 

 

[+bob1200xl]

Do you have the service manual for your 810? 

 

Check the voltages (and the ripple) on the drive.

 

Turn the RPM just a little until you can run an RPM program. Set the RPM to 288. It helps to listen to another drive run and match your 'sound' to that.

 

Once you get the drive to read, you can work on writing.

 

Bob

 

[+Nezgar]

@jaek_3

To help with specifics, theres three "versions" of the 810, which do you have?

- MPI mech without "grass valley board set"

- MPI mech with grass valley board set

- Tandon mech with grass valley board set

 

Hopefully you dont have the first, which is the oldest and most unreliable and temperamental. Quickest way to see is if you have or dont have the  upgrades is if there is PCB above the drive mechanism, which is visible by peering into the front opening.

 

With that said, if the drive appears to be getting past POST, try applying a LITTLE bit of downwards pressure on the pressure on the pressure pad arm when trying to boot/read from a known good disk (ie. freshly formatted/dos files written in a known good drive). If this yields some successful sector reads, adjusting the tension spring may help.

 

Otherwise checking RPM as previously mentioned. Reads are pretty forgiving and will work with RPM off as much as +/- 20RPM - once you get any reading working, you can find an RPM program to narrow it down.