Jump to content
Paul Westphal

1050 US Doubler Speed

Recommended Posts

You could try putting the original 6810 and ROM back in. Have you tried both single and double density writes with the USD in place?

Share this post


Link to post
Share on other sites

yep to writes.

 

the format thing is odd. it spins up to format and the head remains in same place - as if it were formatting, but doesn't move from the same spot.

Share this post


Link to post
Share on other sites

the format thing is odd. it spins up to format and the head remains in same place - as if it were formatting, but doesn't move from the same spot.

 

Sounds like the FDC is not getting the index pulse, which in the case of the 1050 is "faked" from the PIA.

  • Like 2

Share this post


Link to post
Share on other sites

ok, so how to I remedy that? replace FDC, or which other IC(s)?

thanks

Share this post


Link to post
Share on other sites

ok, update:

 

having swapped out all ICs - for working ones and good (mech/stepper) still get exactly the same results using this board

 

latest video:

 

 

 

 

one other thing, the rear metal (aluminium) colour shielding is getting hot

 

 

 

Share this post


Link to post
Share on other sites

check the rpm's of the drive and the alignment.... when is the last time the head was cleaned?

since it boots and steps...it's not all that bad... the heat sink will get pretty darn warm in fact that's normal.... if it truly is getting too hot (Thermal runaway) then you should check the voltages off the regulators attached to the heat sink... get readings while cold and then get readings while hot.... if the voltages vary too greatly or are out of spec... look towards replacing them and other parts of the power supply... capacitors have been known to go bad on the 1050.... in any event.... clean and check rpm are at 288... then move to the other preliminary check as mentioned

Share this post


Link to post
Share on other sites

RPM - not a problem. i've swapped in a known good mech and stepper - same result

 

am convinced this is a board issue. but stumped by it

Share this post


Link to post
Share on other sites

having swapped out all ICs - for working ones and good (mech/stepper) still get exactly the same results using this board

 

You replaced all the ICs? That means you removed the US Doubler altogether and installed back a stock 1050 ROM as well?

 

It could be the board, or a socket, of course. Not the most common failure but certainly not impossible. Would be difficult to diagnose without a scope or a LA at least.

Share this post


Link to post
Share on other sites

Yep and the index pulse circuit, the 1050 of course had no emitter or detector.. It is faked....laying down the first sector and timing off that thereafter....

so u21 and cr10 and the connectors attached correctly seems to be a theme...

Edited by _The Doctor__
  • Like 1

Share this post


Link to post
Share on other sites

Well, I've made my 1st US Doubler RAM pack all by myself after sourcing parts from odd places around the world for a good price on eBay. I had 1 working RAM pack that my dad put together in the early 90's that I've been using recently in my endeavours learning how to burn my own EPROMS (including US Doubler), trying different USD / Stock 1050 code, so this was the next step.

 

Thank you for this very informative thread, Kyle22, MEtalGuy66 for posting pictures of your RAM Stacks, and tep392 for the detailed info about how the stack works, and the tip about connecting pins 14 & 15 instead of running the 2nd wire across the chip. I actually just used the pin I snipped off to place across 14 & 15. And yes I still have to snip off the pin in the top pic, but it works, and looks nice! :thumbsup:

 

Now I can make a few more. :grin:

post-53052-0-64238400-1524202509_thumb.jpg

post-53052-0-39931400-1524202518_thumb.jpg

  • Like 5

Share this post


Link to post
Share on other sites
On 1/7/2014 at 6:55 PM, Kyle22 said:

Disclaimer: I am NOT good at ASCII (or any) Art, so I'll attach a pic. Proceed at your own risk :)

 

 

Take a 6810 RAM. trim off the skinny part of pins 10 and 14.

 

Stack and solder (carefully) all the pins that make contact (except 10 and 14)

 

Using wire-wrap wire, jumper the the TOP pin 10 to the BOTTOM pin 14, and pin 12 of BOTH to TOP pin 14.

post-13040-0-96650900-1389138886_thumb.jpg

post-13040-0-75384400-1389138896_thumb.jpg

post-13040-0-04994800-1389139870_thumb.jpg

The above is the correct wiring.  Follow the text labels, not the seemingly crossed wires.  The wiring here ----> http://www.realdos.net/US Doubler.html  is WRONG and will not work.  Your drive will just sit there like a rock with the power light glowing.  Here's a photo of my rig, after I followed the realdos.net doubler text and then discovered the correct way in this post from Kyle.  The ram in the photo now works, but looks rough because two pins (one I cut accidentally) had to be replaced, and I fiddled with the jumper wiring several times before realizing Kyle was right AND the Realdos doc was WAY wrong in telling me to cut pin 11 on the lower ram chip.  Worth noting, the file I found on Realdos.net, named USDUBLER.rom is the correct one, for disks 1-4.  I haven't tried the second rom there, that ought to be for disks 5-8.  I don't have that many floppy drives man. 🙂  USDUBLER.rom is a standard rom, not a Bob Burner rom, so no editing of header bytes is needed, just load and burn.

 

Nice to have a US Doubler again.  It took a long time to get the parts, used pinball machine parts from Switzerland, and 2732a's from China, but I have a working one now.  The happy drives, the ones that have a track buffer are disconcerting.  With the Doubler I can basically tell you what sector the drive is reading or writing just by the sound of the io and the stepper motor.  The track buffer throws that way off.

 

Best,

 

Jeff

 

20190924_183138.jpg

20190924_183123.jpg

20190924_182928.jpg

Edited by Technoid Mutant
  • Like 4

Share this post


Link to post
Share on other sites

Congrats @Technoid Mutant, I think every 1050 should at least have a US doubler speeder, if only just for true double density support, which the 1050 should have had from the factory.

 

I actually just completed another one of these last night, after also converting a few more 5V 2A DC router power supplies I picked up from Value Village to have XL/XE style DIN-7 connectors. I really like @tep392's method that only requires 1 wire across the top. Then I just use one of the snipped pins for a neat bridge across pins 14 & 15. (But I still like my pictures in post #112 above best)

 

Second_USD_1.thumb.jpg.b89a6b0b1b60727e11aab8c68c9ef2cc.jpg Second_USD_2.thumb.jpg.d6008ca4d9523d5ac95c48eed6258282.jpg
 

This module went into a WST 1050, using slaker's patched ROM that incorporates the faster stepping rate found in the stock WSTR5 ROM. This ROM only works with WST mechs. A happy board in a 1050 with a WST mech is a downgrade for seek performance!

 

2 hours ago, Technoid Mutant said:

The wiring here ----> http://www.realdos.net/US Doubler.html  is WRONG and will not work.

Yeah, this has been noticed before, sorry to hear you wasted time going down that path before seeing this topic.

 

Be careful with some of the US Doubler ROM's floating around out there. I've done quite a bit of investigation into the variants, refer to this post:

https://atariage.com/forums/topic/283132-1050-is-that-an-icd-in-your-socket-or-are-you-happy-to-see-me/?do=findComment&comment=4120404

 

The best "trusted" ROM's to use for Tandon or WST drives:

https://atariage.com/forums/topic/283132-1050-is-that-an-icd-in-your-socket-or-are-you-happy-to-see-me/?do=findComment&comment=4208151

 

The "1-4" and "5-8" versions are merely determined by bytes $13-$16. Just change these bytes from "3241" for 1-4, to "7685" for 5-8. These bytes are not verified by the ROM POST, so you don't have to recalculate a checksum. More here:

https://atariage.com/forums/topic/283132-1050-is-that-an-icd-in-your-socket-or-are-you-happy-to-see-me/?do=findComment&comment=4117858

 

With newer versions of SpartaDOS X supporting up to 15 Drives, you can put drive ID bytes from $31 (Drive 1) all the way to $3F (Drive 15 or "O" in SDX). To test this, I made one EPROM that functions as drive 1, 2, 9 and 10 (J:). One day I want see if I can get 15 1050's running at once this way. :)

 

Edited by Nezgar
more links!
  • Like 2

Share this post


Link to post
Share on other sites
1 hour ago, Nezgar said:

This module went into a WST 1050, using slaker's patched ROM that incorporates the faster stepping rate found in the stock WSTR5 ROM. This ROM only works with WST mechs. A happy board in a 1050 with a WST mech is a downgrade for seek performance!

 

 

I keep waiting for someone to patch a Happy ROM with increased Step Rate for WST Mechs.

Share this post


Link to post
Share on other sites

data:image/gif;base64,R0lGODlhAQABAPABAP///wAAACH5BAEKAAAALAAAAAABAAEAAAICRAEAOw==

 

I have done 15 drives and you need to make an octopus with very short SIO cables... in this way the daisy chain will end up as a series parallel sort of affair not allowing cable capacitance nor signal drop and noise to cause problem....

you can not do a straight daisy chain with them as the signal degrades too much...

 

There was an issue with APE as it responds to some drive ID it shouldn't....

 

and the drive numbers above

Edited by _The Doctor__
  • Like 1
  • Haha 1

Share this post


Link to post
Share on other sites
12 minutes ago, JR> said:

I keep waiting for someone to patch a Happy ROM with increased Step Rate for WST Mechs.

Check out the ROM's in this post: https://atariage.com/forums/topic/213804-1050-happy-warp-with-led-display-v3/?do=findComment&comment=4286826 Faster stepper rate that works on Tandon drives, and also fixed corrupted non-buffered writes with UltraSpeed active. Though not specifically coded for WST, but should be compatible, and comparable.

 

In theory for someone who knows what they're looking at, it's probably only a few bytes to change to adjust these things. The 3-phase stepper sequence is coded in the first few bytes of a US Doubler ROM. Stepper delay is probably a single byte, and the the WST ROM skips the extra '2 steps forward 1 step back' on a forward step, which would require more than a hex-edit methinks...

Share this post


Link to post
Share on other sites

where was the write up on the 2 steps foward 1 step back dance... I couldn't easily search it... but it's got to be in the forum somewhere...

Share this post


Link to post
Share on other sites
18 minutes ago, _The Doctor__ said:

There was an issue with APE as it responds to some drive ID it shouldn't....

Maybe $3F ? Seems to be a Type 2 Poll used by the 850, and for reading the highspeed index from UltraSpeed capable devices...

 

26 minutes ago, _The Doctor__ said:

where was the write up on the 2 steps forward 1 step back dance... I couldn't easily search it... but it's got to be in the forum somewhere...

I think these two posts are a good start:

https://atariage.com/forums/topic/156462-1050-roms/?do=findComment&comment=3581342

https://atariage.com/forums/topic/192618-understanding-atari-floppy-drive-timing/page/2/?tab=comments#comment-2442680

 

Share this post


Link to post
Share on other sites

nice to relive those threads, I re-read the thread with @dhinson919 in it. I will have to read the other.

 

in one of these threads it is stated it does an extra half step in and half step out, that's not two steps in one step out...

that's X steps plus a half in and a half back landing on X. sounds like it's not fully explored as we also get told it does one long half step and one short half step... sounds very odd unless it's in the code to step/pause/step to give timing for other cpu/transmission time since this is happening on seeks... or perhaps step/look at something/step/ for each seek...

Edited by _The Doctor__

Share this post


Link to post
Share on other sites
6 hours ago, Nezgar said:

Congrats @Technoid Mutant, I think every 1050 should at least have a US doubler speeder, if only just for true double density support, which the 1050 should have had from the factory.

 

I actually just completed another one of these last night, after also converting a few more 5V 2A DC router power supplies I picked up from Value Village to have XL/XE style DIN-7 connectors. I really like @tep392's method that only requires 1 wire across the top. Then I just use one of the snipped pins for a neat bridge across pins 14 & 15. (But I still like my pictures in post #112 above best)

 

Second_USD_1.thumb.jpg.b89a6b0b1b60727e11aab8c68c9ef2cc.jpg Second_USD_2.thumb.jpg.d6008ca4d9523d5ac95c48eed6258282.jpg
 

This module went into a WST 1050, using slaker's patched ROM that incorporates the faster stepping rate found in the stock WSTR5 ROM. This ROM only works with WST mechs. A happy board in a 1050 with a WST mech is a downgrade for seek performance!

 

Yeah, this has been noticed before, sorry to hear you wasted time going down that path before seeing this topic.

 

Be careful with some of the US Doubler ROM's floating around out there. I've done quite a bit of investigation into the variants, refer to this post:

https://atariage.com/forums/topic/283132-1050-is-that-an-icd-in-your-socket-or-are-you-happy-to-see-me/?do=findComment&comment=4120404

 

The best "trusted" ROM's to use for Tandon or WST drives:

https://atariage.com/forums/topic/283132-1050-is-that-an-icd-in-your-socket-or-are-you-happy-to-see-me/?do=findComment&comment=4208151

 

The "1-4" and "5-8" versions are merely determined by bytes $13-$16. Just change these bytes from "3241" for 1-4, to "7685" for 5-8. These bytes are not verified by the ROM POST, so you don't have to recalculate a checksum. More here:

https://atariage.com/forums/topic/283132-1050-is-that-an-icd-in-your-socket-or-are-you-happy-to-see-me/?do=findComment&comment=4117858

 

With newer versions of SpartaDOS X supporting up to 15 Drives, you can put drive ID bytes from $31 (Drive 1) all the way to $3F (Drive 15 or "O" in SDX). To test this, I made one EPROM that functions as drive 1, 2, 9 and 10 (J:). One day I want see if I can get 15 1050's running at once this way. :)

 

I would LOVE to see a stack of 15 1050's.  What's that, 2.88meg-ish?  NICE!  Many moons ago, Kenny Sallot ran his Tardis BBS on my Percom?  ATR8000? With a 720k 80-track drives...  With the Percom control block suitably Poked, you could get really fast tracking out of a decent mech.

 

Jeff

 

Share this post


Link to post
Share on other sites

They weren't all doubled drives, I think it weighed in at 2 and a third meg. as the drives were formatted at enhanced and double density, with only the boot drive having DOS and config.sys on it even though only config was of any utility since SDX was in use, everything else was for whatever files you wanted on them.

I think MyDos is what we ended up using as a means to start copying data onto disks for the test, and then realized some issue and the formatted everything with SDX anyway (since that's what we were using in order to access all the drives)... using the sio2pc with APE to fill the drives with 'warez'... that's when we discovered APE had a fit over some drive ID or another...

4 power strips were needed to power the computer monitor and drives...

the drives took up half the table.

Each drive to drive SIO cord was approximately 12 inches (1 foot )

1 foot to the SIO splitter then going to 4 stacks of 3 drives plus one stack of 3 with the sio2pc attached (s2pc eventually removed)

 

Edited by _The Doctor__

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...