Dropcheck Posted May 4, 2015 Share Posted May 4, 2015 Hi, I'm trying to get as much info on this upgrade as I can. Specifically I would anyone who has one to see what kind of switch is being used. Is it a SPDT ie three pins on the bottom of the switch? Or does it have more pins? Thanks, Dropcheck Quote Link to comment Share on other sites More sharing options...
+Larry Posted May 4, 2015 Share Posted May 4, 2015 Mine has 6 pins, but is a 2-way switch. There are 5 wires connected to the switch coming from the usual CSS potted module. There is a 6th wire also coming from the module and it goes to the power switch, so presumably +5VDC. I'll leave this out for a few days if you have other questions. -Larry Edit: In the same storage box, there was my "XF552" drive. Thought I'd take a picture of the board. It's a long board, but there is nothing up front except a few mounting holes. The extra wires are from the power switch and LED that I installed on the front of the drive. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 4, 2015 Author Share Posted May 4, 2015 Mine has 6 pins, but is a 2-way switch. There are 5 wires connected to the switch coming from the usual CSS potted module. There is a 6th wire also coming from the module and it goes to the power switch, so presumably +5VDC. I'll leave this out for a few days if you have other questions. -Larry Edit: In the same storage box, there was my "XF552" drive. Thought I'd take a picture of the board. It's a long board, but there is nothing up front except a few mounting holes. The extra wires are from the power switch and LED that I installed on the front of the drive. Hi Larry, Can you tell me what color wire is going to which pin on the switch? 0 0 0 Center 0 0 0 Quote Link to comment Share on other sites More sharing options...
+Larry Posted May 4, 2015 Share Posted May 4, 2015 red.........orange/blue brown....gray black......n/c Unusual switch -- the toggle barely protrudes from the mounting pad. Quote Link to comment Share on other sites More sharing options...
ricortes Posted May 4, 2015 Share Posted May 4, 2015 (edited) Mine is three pins. Whoops! Make that for your XF Hyper XF. Edited May 4, 2015 by ricortes Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 4, 2015 Author Share Posted May 4, 2015 Thanks Larry, Looks like there is either a second index sensor or he is taking the signal from the original one and logically manipulating inside the module. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 5, 2015 Author Share Posted May 5, 2015 Could the index pulse be faked with a 555 timer and triggered when the MO signal goes high on the 1772? Quote Link to comment Share on other sites More sharing options...
Mathy Posted May 5, 2015 Share Posted May 5, 2015 Hello Lenore Wouldn't it be more reliable to replicate the hardware the XF uses "on the other side"? And easier to use, should the need ever exist to lower the rpm's of the XF? Sincerely Mathy Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 5, 2015 Author Share Posted May 5, 2015 Hello Lenore Wouldn't it be more reliable to replicate the hardware the XF uses "on the other side"? And easier to use, should the need ever exist to lower the rpm's of the XF? Sincerely Mathy Not sure what you mean. A second index sensor? Why would you want to lower the rpm's? Doesn't the OSC compensate for the difference in the industry standard drive mech speed and the rest of the Atari 810/1050 drives? Quote Link to comment Share on other sites More sharing options...
Mathy Posted May 5, 2015 Share Posted May 5, 2015 Hello Lenore Yes, a second index sensor. It does seem to compensate to some extend, but IIRC Silent Service won't run on the XF because of it's higher speed. And what if you (or somebody else) ever wants to add some circuitry to enable the drive to write bad sectors? (BTW please don't get the impression that I know what I am talking about. I'm just repeating what I heard/read and seem to remember) Sincerely Mathy Quote Link to comment Share on other sites More sharing options...
+bob1200xl Posted May 5, 2015 Share Posted May 5, 2015 The 1050 does not use the index sensor at all, so you can flip any disk over and write on the back side. The XF551 does not have that ability - it uses INDEX and writes on both sides without flipping the disk. If you want to flip the disk, you have to feed dummy index pulses to the controller, but only for writing. The higher clock frequency makes the written data fit properly on the disk. When reading, the data is self-clocked, so as long as the VCO can follow the flux transitions, the higher clock frequency (or RPMs) does not matter. I suppose that one could write a routine that measures the time between sector reads in order to copy-protect a disk. That may fail an original disk in the XF551. Bob Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 5, 2015 Author Share Posted May 5, 2015 Hello Lenore Yes, a second index sensor. It does seem to compensate to some extend, but IIRC Silent Service won't run on the XF because of it's higher speed. And what if you (or somebody else) ever wants to add some circuitry to enable the drive to write bad sectors? (BTW please don't get the impression that I know what I am talking about. I'm just repeating what I heard/read and seem to remember) Sincerely Mathy I was trying to stay away from making modifications to the drive itself. In my mind that is counter productive. The drive is probably going to be the first point of failure because of the moving parts. I like the abiltiy to yank the failing drive out and dropping another in and I'm supposedly up and running as before. While I am aware of some commercial software that hiccups on the XF551, that's more a consequence of the copy protection attempt. Much like sometimes you have to unHappy your Happy 810/1050 drive to run some disk based games. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 5, 2015 Author Share Posted May 5, 2015 The 1050 does not use the index sensor at all, so you can flip any disk over and write on the back side. The XF551 does not have that ability - it uses INDEX and writes on both sides without flipping the disk. If you want to flip the disk, you have to feed dummy index pulses to the controller, but only for writing. The higher clock frequency makes the written data fit properly on the disk. When reading, the data is self-clocked, so as long as the VCO can follow the flux transitions, the higher clock frequency (or RPMs) does not matter. I suppose that one could write a routine that measures the time between sector reads in order to copy-protect a disk. That may fail an original disk in the XF551. Bob That may be what the Enhancer is doing. In the instructions it has you cut the IP jumper and connect the module between pin 24 on the WD1772 and the drive mech IP signal at the drive connector. It's also connecting with the Write Protect sensor on the drive mech itself. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 10, 2015 Author Share Posted May 10, 2015 (edited) Why would the module have both an AC source and a TTL power connection?. According to the instructions a white wire is connected to the top right pin of the power on/off switch, a blue wire goes to the +5v pin(7) on U3(74LS38N) and a yellow wire goes to ground pin(14) on U3. Is it being used for timing? 60 Hz Edited May 10, 2015 by Dropcheck Quote Link to comment Share on other sites More sharing options...
1050 Posted May 11, 2015 Share Posted May 11, 2015 Instead of 555 timer use a divide by ten on 60 Hz and you have the missing index pulse for 360 RPM flipped floppies. If so, with white wire disconnected the drive should loose it's ability to format back side of flippies, if it has that ability to begin with. Don't have one to know this much, just tossing out ideas. I won't be paying for damages incurred either, standard disclaimer applies. 5 volt zener and resistor would limit 9 volt to 5 for signal use within the module. 1 Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 11, 2015 Author Share Posted May 11, 2015 Interesting. So if the XF551 drive rotation is 300RPM and you use a divde by ten for the 60HZ you would have the missing index pulse? Quote Link to comment Share on other sites More sharing options...
Joey Z Posted May 11, 2015 Share Posted May 11, 2015 Interesting. So if the XF551 drive rotation is 300RPM and you use a divde by ten for the 60HZ you would have the missing index pulse? not for 300RPM, he said for 360RPM. 360RPM is 6RPS, and 300RPM is 5RPS. 60/10 is 6hz corresponding to 6RPS. for 5RPS, you would need divide by twelve. Quote Link to comment Share on other sites More sharing options...
+bob1200xl Posted May 11, 2015 Share Posted May 11, 2015 It doesn't really matter what the RPM is. What you need is a pulse that occurs at a regular interval to act as an Index pulse. It is only required during Format. You send the controller a Format command and it waits for an Index pulse to start the Format sequence. You can wait two revolutions of the disk or 1/10 of a revolution - all the same to the controller. He doesn't care. That said, some of the other commands may look at Index - things like timeouts and such. I don't think the XF551 uses any of them, though. You can read and write to the back side of a disk with a stock XF551 - you just can't Format. Bob Quote Link to comment Share on other sites More sharing options...
+CharlieChaplin Posted May 11, 2015 Share Posted May 11, 2015 (edited) Well, this is true for Mitsumi drives in the XF551 (most common it seems). But here in Europe Atari also used Chinon drives (well-known from Amiga Computers) and errm, they cannot read, write, format the backside of a disk without an index hole or without an index pulse. Insert a disk into a Chinon drive without a second index hole and it will not stop spinning/rotating... [q] You can read and write to the back side of a disk with a stock XF551 - you just can't Format. Bob [\q] Edited May 11, 2015 by CharlieChaplin Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 11, 2015 Author Share Posted May 11, 2015 The whole point of this is to figure out how CSS managed this trick. They went to great lengths to hide their technique and when they were a going company and stood to lose sales that was a valid move. Now however the website is there for posterity's sake and has not been updated for nearly 15 years. No production or sales are being generated. It appears to have been abandoned. This info is in danger of being lost, if it hasn't been already. I have tried to go the honorable route and broker a deal for the information on this as well as other XF551 mods that CSS produced, but after a few positive responses Bob Puff of CSS ceased to respond to me emails. At no point did he say no. For all I know he simply did not have info to give me. I don't want the credit for this trick, I didn't invent it. The credit remains with the company that produced this mod, CSS. Please help me save this and the other XF551 mods by CSS from the black hole of time. Who knows maybe we can move on to the other CSS products in time. 3 Quote Link to comment Share on other sites More sharing options...
Mathy Posted May 12, 2015 Share Posted May 12, 2015 (edited) Hello Lenore I just found this on my Imac. It's not mine and I'm not sure if it helps. Sincerely Mathy 60Hz pulse for XF551 Enhancer.pdf Edited May 12, 2015 by Mathy Quote Link to comment Share on other sites More sharing options...
1050 Posted May 12, 2015 Share Posted May 12, 2015 Joey Z covered my senior moment where I was thinking they turned at 360 instead of 300 RPM. Cough, cough, ahemm. Yes, divided by 12, 60 Hz will give 5 Hz which is the same rate as the index pulse needed. And for PAL version divide by 10 for the same 5 Hz signal. As Bob suggests an exact pulse replication isn't needed, just a close to right pulse will start the track formatting. Andreas brings up another issue that might make us want a spot on signal though, what is it with the chinon drive? I've looked for jumper settings before and some are vast in the complex interdependent manner some work. Nothing to be found on the F-502 chinon though that contains what jumpers 6 thru 9 might be doing. Not too sure about 10 (360 silkscreen) much less 11 (360/300 silkscreen). It might be best to derive a true 5 Hz signal? 5.25-jumpers.pdf red.........orange/blue brown....gray black......n/c The left side of the switch does pulse substitution and the other deals with defeating the write protect system. Brown would be the index pulse in to the FDC, pull up resistor R22 remains connected to FDC IP/ pin throughout, but the trace from that junction going to drive connector should be cut. Red would attach to this trace past the cut to restore normal IP/ signal. Black would be the generated 5 Hz signal pulse. On the right side of the switch above, either color could be FDC WP/ pin at pull up resistor R24 and this resistor also remains connected to the FDC WP/ pin throughout. Again the trace from that junction to the drive connector is cut with the other color wire making the cut trace restored when this side of the switch is closed. In the open position, this effectively removes the drive's WP/ sensor and it can no longer bring the signal down to write protect the system. Two switches might be nicer giving more control, but it's hardly an issue to be considered. Save the price of a switch and just use one? I would actually use a 4.2 volt zener and resistor to be determined since TTL logic doesn't go rail to rail. Best to avoid possible lockups at inputs above 5+, 4.2 zener should give us that freedom from latch up troubles and still work just fine. Start with a 10k buffer resistor between the zener and power supply at the switch to measure if the zener is at voltage. Lower resistance until it is and then shunt the zener with a 420 ohm resistor to simulate a 10 ma load thru the zener. Lower the buffer resistor more until voltage is again at 4.2 and then remove the shunt resistor such that 10 ma flows thru the zener so that it is operating on it's 'knee' well and good. This will protect circuits down stream. Should plan on at least a 500 milliwatt zener and you can go bigger. 74LS92 is a divide by 12 counter, feed it a 60 Hz clock at CKA (pin 14), tie QA (pin 12) to CKB (pin 1) and get a close to 50% duty cycle clock at 5 Hz from QD (pin eight). I know it looks like O but it's a Q. Pins 6 and 7 are grounded. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 12, 2015 Author Share Posted May 12, 2015 Hello Lenore I just found this on my Imac. It's not mine and I'm not sure if it helps. Sincerely Mathy Hi Mathy, Thanks for the info. It appears this is a mod to the mod. It adds an on/off LED and probably a more stable cleaner index pulse to the AC input white wire to the module. Definitely an inprovement for the LED alone. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 12, 2015 Author Share Posted May 12, 2015 Here's what I have so far not counting Mathy's post. I am working to implement the LED mod and 555 timer. Will also look at how 1050's post fits in too and update. This is the connection based on the XF551 Enhancer install manual instructions. Quote Link to comment Share on other sites More sharing options...
Dropcheck Posted May 12, 2015 Author Share Posted May 12, 2015 So why the NANDed MO signal? Quote Link to comment Share on other sites More sharing options...
Like us on Facebook
Follow us on Twitter
AtariAge on Instagram
Watch us on YouTube
Chat on Discord
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.