Fabricating Timeline Micro Lynx transport cables

So sorry to hear the news about pianodano. He seemed from what he wrote in the threads I've read so far to have been a "been there, willing to share" individual, who also encouraged folks without undue condescension, or other BS. I would have enjoyed interacting with him in these current matters, and I guess the few of us willing to try the route of DAW "Clock Only Master" ( which brought me here in the first place) owe him a debt of gratitude for paving the way. My prayers and condolences go out to his wife and loved ones, of course. R.I.P. Danny, and may the keys you now tickle awe the harps to silence!
 
WOW! sweetbeats, that file set you laid on us knocks me out! Did you get all those from the Session Control website before they shut down? Holy smokes, Service Bulletins galore, maybe even a slightly larger set of cable drawings (haven't done a file count yet) than previous, sales sheets of Option Cards, Install Instructions- the works. I've downloaded everything I could, so if there are any future needs for those files I'll try my best to pass them on. And the extensive photo spread of the System Unit and Option cards- once again WOW!

Now all the group needs is a Service Manual, and we'll be ready to resurrect any Microlynx we come across!
One last thing before you button up the unit again, if you don't mind- could you write down somewhere what the PCB Revision numbers are so that we could verify EPROM levels to PCB levels?
Some of your photos do show the PCB Revs, but this would be a very helpful link in the chain of support, so to speak.

So I just have done some preliminary investigation on the first part I wanted to identify- U19 on the motherboard. It turns to be (as I suspected) a type of EPROM, however not the EEPROM I thought it might be. It is a Xilinx XC1736DPC 36k-bit serially-programmed EPROM IC. It may be a One Time Programmable, but I have to check the datasheet more closely to be sure. What they do mention is its use in conjunction with Field Programmable Gate Arrays (which act like advanced logic circuits). I'll have to explore the square multi-pin packages on the motherboard more closely to try to figure out what they are.
For now, as the Xilinx PDF I first examined called out the whole XC1700 family as obsolete, we should collectively put it on the endangered list, and attempt to get a read of the code at the same time as the main 27C512's are read back and saved. I just saw a couple of pieces listed on fleabay, so they aren't gone completely yet, but Digikey couldn't find a match, so we should tred carefully, but swiftly.

About the IC's that are socketed that you couldn't see clearly (the yellowish colored ones buried beneath an Option Card), they may be resistor arrays, which are often used as low current pull-up/pull-down resistors for data lines etc., but I'll have to examine more closely to be sure (can't quite read ID #'s off the photo due to glare).

That's a great shot of the 3.6V Hitachi backup battery, I'll have to look it up for possible replaceability.

About the request for forum moderator help in a re-name of the thread, I'll bow to you folks of greater wisdom and longer site participation, but I was planning on starting a new thread just for documenting my progress (as you have done so many times here sweetbeats) on the EPROM, and Microlynx knowledgebase. Too bad we couldn't group all the Microlynx stuff into a sub-forum or topic heading or something, as right now I don't see its equivalent anywhere on the web. We may soon be creating the most complete knowledge base on the planet for this unit, outside of Bill McMeekin's file cabinet, wherever he and it reside.
Anyway, thanks again to all you folks sweetbeats, timkroeger, miroslav, (and of course the late-great pianodano) for giving me an excellent starting point and apprenticeship with this very useful product!
 
OK, sweetbeats, -I just looked up the part data for the yellow, socketed 16 pin DIP IC's (one of which is print ID'd on the daughter PCB of the M3 Machine Expansion Card as U24), they are:
Toshiba #TLP627-4 quad opto-isolators, and are probably used in this case as a means of controlling some of the data lines of the Transport Port, so that there is no direct galvanic (current flow-path) link to the more sensitive logic IC's. Much the same as a well designed MIDI port using opto-isolators doesn't let outside current flow directly into deeper circuitry, such as the UART chip itself. A sort of protective buffer/fuse, if you will. At least that's my understanding of it, although there may be other uses for them.
 
About the request for forum moderator help in a re-name of the thread, I'll bow to you folks of greater wisdom and longer site participation, but I was planning on starting a new thread just for documenting my progress (as you have done so many times here sweetbeats) on the EPROM, and Microlynx knowledgebase!

I think that would be the simplest thing...if you start a new thread, just add links to these existing threads in the fist post, even if you copy/duplicate some of the info from them, into the new thread.
That way those threads can also continue as they are...and they would at the same time be linked into a new documentation thread.
I don't think you have to go nuts about organizing and categorizing everything. Any newcomers will benefit more by having to read through it all...the journey is where things are learned. :)
 
Interesting. Is the MicroLynx actually using an FPGA? Or are they just using the companion EPROM for some weird purpose? Looking at the datasheet it seems to bang out a 1-bit datastream intended for consumption by the FPGA. I'm not sure what else you'd do with that that couldn't be achieved more easily with a normal 8-bit EPROM, but I'm sure there's something.

AFAIK FPGA devices are usually used for prototyping because it's cheaper to mask an ASIC (e.g. the Fostex custom CPU) unless you're doing a very small run of very expensive units, or you desperately need to be able to update the logic in future.

EDIT: Sorry, just seen the photos. Yes, that's using an FPGA alright. Interesting!
 
Yes jpmorris, there are apparently a few FPGA's involved per unit, I haven't had the time to research every one of the square 80+ lead chip in the units, but for sure those Xilinx ones are, and those are what the (very specifically designed as FPGA "feeders") XC1736DPC's are for.

So again (I speculate that) there will be one of those required per transport/machine control port (as the M3 Option card has one, I believe). I just read a white paper outline from Xilinx about utilizing a FPGA to do advanced digital PLL duties, and this would make sense, as the 80-bit SMPTE stream from each source deck would have to be serially stored, compared, and time offset somehow - unless you can think of a different use for them?
Ever popped the hood on your Lynx 1 to look at the chipset? I wouldn't be too surprised to find something similar going on.
My unit hasn't arrived yet, but when it does I'll do some more thorough research. Sounds like we're developing a basic Theory Of Operation, and that's a good and useful thing for troubleshooting.
 
Yes jpmorris, there are apparently a few FPGA's involved per unit, I haven't had the time to research every one of the square 80+ lead chip in the units, but for sure those Xilinx ones are, and those are what the (very specifically designed as FPGA "feeders") XC1736DPC's are for.

So again (I speculate that) there will be one of those required per transport/machine control port (as the M3 Option card has one, I believe). I just read a white paper outline from Xilinx about utilizing a FPGA to do advanced digital PLL duties, and this would make sense, as the 80-bit SMPTE stream from each source deck would have to be serially stored, compared, and time offset somehow - unless you can think of a different use for them?
Ever popped the hood on your Lynx 1 to look at the chipset? I wouldn't be too surprised to find something similar going on.

Heh, I would - 1985 is a bit too early for that kind of thing. A ULA or PAL device is more likely. It's been a couple of years since I copied the EPROM, but it looked fairly unremarkable - an early 80s microcomputer with a daughterboard for the pilottone stuff. What did surprise me was the lack of battery backup, but it turns out the A807 has an EEPROM to store its settings and that was only two years afterwards.
 
Of course you're right about the specific IC's or even IC families being different. After all, the concept of Application Specific IC's was fairly new for the '80's, such that the Uncommited Logic Arrays and Programmable Logic Arrays you mentioned as being possibly involved were "best in class" for that era.

The overall concept of wide-word parallel to serial conversion for the purpose of quasi-realtime Phase Lock Looping is what I was referring to as similar between the Microlynx and Lynx 1/ Lynx 2 units.

Do you still have the .bin files from your EPROM refresh/copy somewhere? Might be cool to post those up for posterity's sake, as I'm going to try to do just that for my Microlynx when it arrives (and I can get a decent burner purchased).
 
OK, I got curious and took one of them apart. I'd forgotten what a forest of 7400 logic chips it had inside. Much like an early 80s microcomputer.
The CPU is a Siemens SAB8032AP, with an Intel P8256 peripheral IO chip. Between those and the EPROM are two MAXIM ICM7218 controllers for the 7-segment display - I'd hazard a guess it's 4 digits each chip but I'm a little surprised they didn't drive the display directly from the CPU.
The NVRAM is a Dallas Semiconductor DS1220AB static RAM chip which bizarrely can act as a drop-in replacement for either a 2K RAM chip, a 2716 EPROM or a 2816 EEPROM.

With the daughterboard removed, we have yet more 7400 and 4000 logic gates, and an Analog Devices AD667 12-bit DAC. I'd hazard a wild guess that the MK2 conglomerates the forest of logic gates into a ULA or similar device before going FPGA in the MicroLynx.


Lynx1-annotated40.jpgLynx1-no-daughterboard33.jpg

EDIT: Yes, I have two versions of the ROM, 4.10 and 5.00. As part of that project I switched both machines to 4.10, since the 5.00 ROM appears to have video editing extensions and goes into some kind of editor config mode each time it boots, which got annoying. The ROM images are in the Motorola S-record format that the ART programmer software uses by default so you'd need to convert them to see the raw data.
 
Nice comprehensive post there jpmorris. Especially interesting is the commonality between the first generation of TimeLine products, and the later ones, aside from the increased PCB density that the FPGA's gave to the Microlynx.

Just a quick example - the Analog Devices D/A converter is the same one that is present in the Microlynx.
I'm going to hazard a guess that it is being used for the actual analog voltage required to slew the motors of the ATRs it is controlling, but I can't be sure yet.
Edit: Yes, there appear to be one for every transport control port, so I think this guess is correct - and thanks again to sweetbeats for providing the higher-rez photos on page 4 to help me verify it.

Maybe you already have a solution, but if you're ever so inclined to convert your SREC files to binary for posterity posting, I just came across this:
Confidentialite- France
The file name result of the sourceforge project is srec_cat.exe
I haven't used it yet (gotta get a burner first!) but it was recommended to investigate by keil-dot-com as a replacement of the older DOS based programmer-file converters like MOT2BIN.exe
 
Nice comprehensive post there jpmorris. Especially interesting is the commonality between the first generation of TimeLine products, and the later ones, aside from the increased PCB density that the FPGA's gave to the Microlynx.

Just a quick example - the Analog Devices D/A converter is the same one that is present in the Microlynx.
I'm going to hazard a guess that it is being used for the actual analog voltage required to slew the motors of the ATRs it is controlling, but I can't be sure yet.

I hadn't thought of that, but it does make sense. As for the ROM images, I found MOT2BIN myself and ran it in DOSbox. Which I use for games mostly, but it also works well to run the EPROM programmer's DOS-based software.
 
Gotcha jpmorris, I hadn't realized the ART programmer you mentioned was a DOS-based, presumably parallel port variety. Still might be useful to post the .bin files somewhere as a backup, if you have the time or inclination, as I'm not sure whether the modern (USB) burners "talk" SREC or not, at least without further translation.
 
So, I better "swing the ship around" a minute and post the following questions to the group (and simultaneously get this thread back on-topic, which is long overdue).
Miroslav pointed out the following fleabay item, thinking it might help me get a head start on putting together my cable for connecting my Microlynx (arriving tomorrow, UPS says) to the AI-2 (just received today, preceded by the AI-2 Service Manual). The item is:

TimeLine LYNX-2 to Alesis ADAT Synchronizer cable 20' | eBay

The problem I'm having, given that there's only one picture, is believing that the cable is genuine for the purpose of syncing to a Lynx-2. The end with the 1/4" connector (I think it's for Timecode Send) is I believe the appropriate one for the Lynx-2 end, and the one for the AI-2 appears to be a 25 pin D-sub, with most of its pins removed (as expected, given that there are only six RS-422 data lines required, plus a small handful of other signals spoken of in the AI-2 manual). Without seeing the alleged Lynx-2 end "head-on", I can't tell the size and pin count of the connector.

The Lynx-2 Transport Port is on a 50-pin D-sub, which I could believe is roughly the same size (in width of the connector hood) as the AI-2's. After just going over the pinout of the port from the manual (Table 10-8, pg. 10-31), it doesn't appear to carry any RS-422 at all- in fact the only RS-422 ports on the Lynx-2 are carried by much narrower-width 9 pin D-subs. Have any of you ever seen a 9 pin D-sub in a larger hood?

I'm going to ask the seller to please provide pictures of the two D-subs side-by-side, and ask that they try to take a picture of some length of the cable's typical print ID, to try and suss this out. Can anyone recommend any other questions to ask that might help ID the cable absolutely? Does anyone have an original TimeLine cable they could photograph that might show where they print the part number?
Thanks again.
 
Unless I'm much mistaken it works like this - one Lynx can manage one deck. To synchronise two decks, you need one Lynx for each, and the RS422 lines are to get the pair to communicate with each other. You can chain a bunch of them together to involve multiple tape decks and lock them all to videotape etc. I don't believe the serial lines are ever used for talking to a deck, just between Lynxes.

Truth be told I never managed to get mine to work properly, and I use them now as handy timecode displays and for cleaning up incoming timecodes. For the actual syncing I use an ATS-500.
 
Yes, I realize that the Lynx-1 and Lynx-2 only support one transport at a time. The Microlynx actually has transport and RS-422 combined onto one connector (the 40 pin JAE-style). The AI-2 sort of acts as an intermediary between an ADAT group and the Lynx-2 or Microlynx systems, so no "traditional" transport control as an ATR requires, with motor control, etc..
So the intended Lynx-2 (or Microlynx) to AI-2 cable has to carry RS-422, some sort of Ref signal (like Wordclock), and possibly a separate SMPTE signal. We're talking maybe 10 or so conductors in the cable (if each signal has to carry a companion ground line).
Do you still have any cables for your Lynx-1 that might have an ID or TimeLine part number printed on them somewhere, -probably would be a number like 70CXXX?
 
Wait- do you have an IF-500 to go with your Tascam ATS-500?
I don't know those systems well enough to know if you needed an IF-500, like the MTS-1000 did with the IF-1000 in order to get transport control of a (or multiple) ATR's. If so, do you have a pinout for the cable that goes between the ATS-500 and IF-500, as I am told Tascam used (but never called out a part number in the MTS-1000 Manual) the same cable for their later series sync "pairs"?
My other option should this experiment with the Microlynx fail, is to try to get an IF-1000 to go with my MTS-1000, and at least run my Otari and MSR-16 together from the Tascam sync units.
 
Wait- do you have an IF-500 to go with your Tascam ATS-500?
I don't know those systems well enough to know if you needed an IF-500, like the MTS-1000 did with the IF-1000 in order to get transport control of a (or multiple) ATR's. If so, do you have a pinout for the cable that goes between the ATS-500 and IF-500, as I am told Tascam used (but never called out a part number in the MTS-1000 Manual) the same cable for their later series sync "pairs"?
My other option should this experiment with the Microlynx fail, is to try to get an IF-1000 to go with my MTS-1000, and at least run my Otari and MSR-16 together from the Tascam sync units.

I don't have the parallel interface or cables, unfortunately.

The original setup was locking two TSR-8s together, and the Lynx Mk1 didn't support TASCAM's serial protocol, at least not with either firmware version I have. I built some cables to try and make it work using the parallel (Accessory 1) port and telling the Lynx it was an 85-16, or Tascam 58 or something, but I never managed to get that setup going. With the ATS-500, it was using the Accessory-2 serial port, and after a bit of trial, error and realising that the schematics were wrong in the manual, I was able to make a set of cables to lock both machines together.

Eventually this became a hassle so I got an MSR-24, and got into a routine of recording the vocals and bass and anything else that needs a lot of punch-ins on the TSR-8, since the parts are cheaper. I record a MIDI guide track on the TSR-8, and when I'm happy I use the ATS-500 to lock the two machines together and transfer the vocals to the 24-track. The MSR-24 uses the same protocol which made things easier.

Finally I got an MX80 to replace the MSR-24. The Otari doesn't speak TASCAM's serial protocol, but the ATS500 it can do a code-only lock, so I roll the Otari first, and then get the TSR-8 to chase that.
IIRC the Lynx does support the MX70/MX80 so I could probably do this with the TASCAM as master, but what I have right now is working well enough so I've not really thought about changing it.

If you're using an MSR-16 and MTS-1000 the same option should apply - driving the MSR-16 direct via serial and locking it to an incoming timecode from the Otari.

It's worth noting that if you do the MSR-16 slave method, the capstan control line (i.e. the bit missing from the schematics) is supposed to be shielded. Mine wasn't, and as a result some of the values the ATS500 is showing are a bit unusual but it does work. Someone else had complete failure with the same setup until they had their cable remade with a shielded line for the capstan control. I may have got lucky.
 
Thanks for the tips there, I've considered a few of the options you've spoken of, and I'll consider them as a fallback plan if the current line of experimentation should fail. I'll also note that its quite illuminating to see your progression of "workflow evolution". I'll have to feel out the best options for mine when I can get there, but doubtless there will be modifications along the way, and it's encouraging to see which options you've taken to get you where you are currently.

The basic notion is that because I'll be attempting to integrate three, possibly four systems together I wanted to try the idea of slaving two ATR's and 3 DTR's together through a combination of latter-era sync devices, along with a secondary DAW. I would then be utilizing a primary DAW as a "Code Only Master", with the "house sync" clock source being something like a Lucid SSG-192 (as miroslav has achieved) that would potentially output Black & Burst for the ATR and DTR sub-systems, and two flavors of wordclock for the digital DAWs- 48kHz & 96kHz.

In essence every sub-system could be utilized on its own, in pairs, or as a "supergroup" if the required project track count required it (or circumstances caused a problem with any one system). Oh, and I think I'd be fine with the various differing levels of audio quality, from lower track width analog to lower sample rate (48kHz) and bit depth (20bit) up to the top of the heap 24bit/96kHz of the primary DAW - it would just come down to prioritizing and selecting the medium to suit the intended purpose and focus of the tracks in the mix.

I've also read a bit of others' experiences with needing the ATR Transport Cables to have certain lines well shielded (as well as having logical pairings be made up of twisted-pair wiring) to keep the crosstalk to levels that avoid sync dropouts due to noise. Thanks for reminding me of that detail.
 
The Missing Data Is Beginning To Take Shape!

OK folks, so after writing up a request for the added photo from the seller of the Lynx-2 to AI-2 cable, he responded immediately with "Done. Cheers." I've attached the images below. Now I just need to do a pin interpretation of the photo, finish my comparison of the Lynx-2 to Microlynx pinout, verify that result against the Alesis schemos, then I'll be ready to order the parts to complete the first cable.

It is indeed strange that TimeLine would have used the 50 pin Transport Port, as I have explained that to my knowledge the Alesis would need the RS-422 data lines, and for the Lynx-2 there were no such signals named on the port. Perhaps they were "faking it" by utilizing some pins that were usually assigned to some other function, and changing the meaning of the data lines virtually, i.e. in software.

Also, my hunch was basically correct about them using approximately 10 lines for the signals. There appear to be 11 active pins on the cable. Two of the missing pins on the Lynx-2 end are simply present on the 1/4" TRS "pigtail". ;D

Lynx 2 to Alesis AI-2 (s-l1600).jpg
s-l1600B.jpg
 
Well folks, I just spent the last couple of hours recovering from a real "DOH!" moment. In my rapid-paced research of the last couple of days, I somehow didn't realize I had grabbed data from the wrong freaking pinout chart in the Lynx-2 Manual!

There are apparently two of them, one (the appropriate one for "normal" Lynx-2 units) appears as "Table Chapter 3-4" on page 3-12, and the other one appears as "Table Chapter 10-8" on page 10-31. The table in chapter 10 deals only with a special film resolving option card. The outputs of it are remarkably similar, which is why I didn't notice it right away, but it just kept bugging me how the cable on fleabay was supposed to control the AI-2 without a breakout connector, or dedicated pins for the RS-422 data lines. I kept seeing references to Biphase signals, and thought maybe they were using the terms in an odd way to refer to the Transport Control Voltages.

Anyway, I'm back on track, and am more convinced I should pony up for the fleabay cable, if for no other reason than to verify with 100% certainty that the pinouts are correct. The good news is that with the correction of my error, I can see that it will be much easier to make cables that translate from the Lynx-2 to the Microlynx, and vice-versa. Although it's not as fun as looking at photos, I thought I'd share with the group a couple of .gif translations of the cable pinouts, complete with blacked-out indications of the missing pins, and my document which compares the pinouts for the Lynx-2 and Microlynx.....a sort-of Rosetta Stone of this cable-making process. Yeah, right, that's what we are....Techno-Archeologists!:facepalm:

First, the Lynx-2 end of the cable (as you are facing the pins):

TimeLine Lynx-2 End.gif


Next the Alesis AI-2 end:

Alesis AI-2 End.gif

Yes, they won't win any art awards, but they gave me the practical info to produce the following (because they helped point in a BIG way to the fact that there was an error in the data). Let's hope I can post these as a .rtf file, as I don't have a PDF "burner" installed at the moment.
 
Back
Top