Topic: Coax-in mod for JB3 (?)

[Lee]

I was bored tonight and took my JB3 apart just for the helluvit (such is life).  Anyway, I got to thinking- would there be a way to take the optical/analog jack out and put an AD-20 style coax jack in its place.  Obviously you'd lose the ability to input analog, but wouldn't it be worth it to not have to deal with optical cables?  I doubt there's a pin-for-pin replacement, but if the voltage to the optical transmitter is just standard S/PDIF (didn't I read this somewhere?) it might be possible to jump it directly to a wire.  Here's some pictures of the inside of it...

Top of board, above the jack


Top, again


close up of the jack, from the bottom of the board


bottom, again


top of board, wider angle


techno-geeks get to work (please)

[Kindguy]

This would be great for the stealthers. I couldn't imaging having to stealth with an optical cable.

I guess most go line in.

[Brian Skalinder]

Doug made a brief post over on the Oade board about this a while ago:

http://www.oade.com/Tapers_Section/Forum/dcboard.php?az=show_mesg&forum=105&topic_id=134&mesg_id=142&page=

That might not be so easy. First you need to be able to install a chip that will take the S/PDIF input level of 0.5 to 1.0 volts and translate it to the TTL logic level the JB3 uses, probably 3.3 volts for a logic 1. If you know you will be using, say an SBM-1 for an A/D, you can use the Sony TTL logic level output of the 7 pin port and not need to use a chip to translate the 0.5 to 1.0 volts S/PDIF signal. You would only need to remove the optical receiver and add a coax connector with its output tied to the same point on the JB3 PCB as the output of the receiver was connected. Use a polypropylene capacitor to decouple the input and do not forget to ground the coax input. Some A/D converters are easy to modify and add a TTL logic level output, so this is another option to the SBM and will no doubt be easier than trying to make the JB3 take a true S/PDIF input. Alternately you can buy a low cost ($20-25) coax to optical converter to allow you to receive a coax input by changing to an optical output. If you use a splitter with the coax input of the low cost format converter you can also pass a signal to the next deck in line.
I hope this helps...Doug

[twatts (pants are so over-rated...)]

Doug made a brief post over on the Oade board about this a while ago:

http://www.oade.com/Tapers_Section/Forum/dcboard.php?az=show_mesg&forum=105&topic_id=134&mesg_id=142&page=

That might not be so easy. First you need to be able to install a chip that will take the S/PDIF input level of 0.5 to 1.0 volts and translate it to the TTL logic level the JB3 uses, probably 3.3 volts for a logic 1. If you know you will be using, say an SBM-1 for an A/D, you can use the Sony TTL logic level output of the 7 pin port and not need to use a chip to translate the 0.5 to 1.0 volts S/PDIF signal. You would only need to remove the optical receiver and add a coax connector with its output tied to the same point on the JB3 PCB as the output of the receiver was connected. Use a polypropylene capacitor to decouple the input and do not forget to ground the coax input. Some A/D converters are easy to modify and add a TTL logic level output, so this is another option to the SBM and will no doubt be easier than trying to make the JB3 take a true S/PDIF input. Alternately you can buy a low cost ($20-25) coax to optical converter to allow you to receive a coax input by changing to an optical output. If you use a splitter with the coax input of the low cost format converter you can also pass a signal to the next deck in line.
I hope this helps...Doug

What does this mean in Moron English...  Is this something that the Oade's may pursue?  Sounds like something they aren't too interested in following; of course he can get a converter, duh!  That's not the point of the mod.  Its to lose the optical!

I wonder if he could add the TTL mod to the UA5???

Terry

[Todd R]

Ok, I'll attempt english, but it might turn out to be techno-babble, who knows.

The true SPDIF standard is -0.5volts for a logic 0 and +0.5volts for a logic 1.  Most gear doesn't really follow the SPDIF standard exactly (probably since the spdif standard is assinine).  So A/D gear will put out 0volts for a logic 0, and 2-5 volts for a logic 1.

Semiconductor chips (ICs) in general require "TTL" logic levels--transistor-to-transistor logic.  Theoretically, TTL levels are 0volts for a logic 0 and 5volts for a logic 1.  In practice, most ICs require somewhere 3.2 and 3.6 volts to recognize a signal as being high enough to be called a logic 1 (this is the 3.3volts Doug is talking about).

The trouble is the voltage level of the digital signal coming over the coax from the A/D converter might be high enough to meet or exceed the spdif standard, but it is not high enough to meet the TTL standard required by the ICs onboard the JB3.  This is the case, I believe, with my V3.  Apparently according to Doug, the Sony SBM1 puts out a TTL level signal on its 7-pin out (meaning, more than ~3.3volts for a logic 1, meaning alternatively that Sony does not conform to the spdif--Sony/Philips Digital InterFace--since they know the spdif standard is moronic).

So if you're using a SBM1, you could replace the optical jack with a 1/8" jack used for a coax input.  Tie this right to where the old optical went on the board (well, with the de-coupling capacitor Doug notes on the hot lead) and all would work fine.  But then when you upgrade from the SBM1 to a V3, it would not work since the V3 doesn't meet TTL levels.

You can get buffering chips that will take lower than TTL levels and boost them to TTL levels.  A good coax-to-optical converter design for example would do this.  I imagine this is what is done in say a Hosa coax>optical converter.  The crappy infamous "greybox" converter (which I bought and have lying around somewhere) does not use these buffering chips.  Which is why the greybox converter does not work with the V3.

Doug also says that rather that trying to install the buffering chips onboard the JB3 so it can take any coax output from an A/D converter, it might be easier to modify your A/D converter to be sure it sends out TTL level signals over its coax out.  The trouble with this approach is that if you change outboard A/D converters, your mod'd JB3 may stop taking the coax signal.

Bottom line to all this is that it just won't be too easy since the most flexible way will be to do all the changes on the JB3, but this will require adding circuitry to an existing surface-mount board.  I've thought numerous times of trying to attack it, but I don't want to mess with my JB3 and I haven't gotten motivated enough to buy a second JB3 that I can f*ck with and probably destroy.

-Todd

[twatts (pants are so over-rated...)]

Ok, I'll attempt english, but it might turn out to be techno-babble, who knows.

The true SPDIF standard is -0.5volts for a logic 0 and +0.5volts for a logic 1.  Most gear doesn't really follow the SPDIF standard exactly (probably since the spdif standard is assinine).  So A/D gear will put out 0volts for a logic 0, and 2-5 volts for a logic 1.

Semiconductor chips (ICs) in general require "TTL" logic levels--transistor-to-transistor logic.  Theoretically, TTL levels are 0volts for a logic 0 and 5volts for a logic 1.  In practice, most ICs require somewhere 3.2 and 3.6 volts to recognize a signal as being high enough to be called a logic 1 (this is the 3.3volts Doug is talking about).

The trouble is the voltage level of the digital signal coming over the coax from the A/D converter might be high enough to meet or exceed the spdif standard, but it is not high enough to meet the TTL standard required by the ICs onboard the JB3.  This is the case, I believe, with my V3.  Apparently according to Doug, the Sony SBM1 puts out a TTL level signal on its 7-pin out (meaning, more than ~3.3volts for a logic 1, meaning alternatively that Sony does not conform to the spdif--Sony/Philips Digital InterFace--since they know the spdif standard is moronic).

So if you're using a SBM1, you could replace the optical jack with a 1/8" jack used for a coax input.  Tie this right to where the old optical went on the board (well, with the de-coupling capacitor Doug notes on the hot lead) and all would work fine.  But then when you upgrade from the SBM1 to a V3, it would not work since the V3 doesn't meet TTL levels.

You can get buffering chips that will take lower than TTL levels and boost them to TTL levels.  A good coax-to-optical converter design for example would do this.  I imagine this is what is done in say a Hosa coax>optical converter.  The crappy infamous "greybox" converter (which I bought and have lying around somewhere) does not use these buffering chips.  Which is why the greybox converter does not work with the V3.

Doug also says that rather that trying to install the buffering chips onboard the JB3 so it can take any coax output from an A/D converter, it might be easier to modify your A/D converter to be sure it sends out TTL level signals over its coax out.  The trouble with this approach is that if you change outboard A/D converters, your mod'd JB3 may stop taking the coax signal.

Bottom line to all this is that it just won't be too easy since the most flexible way will be to do all the changes on the JB3, but this will require adding circuitry to an existing surface-mount board.  I've thought numerous times of trying to attack it, but I don't want to mess with my JB3 and I haven't gotten motivated enough to buy a second JB3 that I can f*ck with and probably destroy.

-Todd

+T, you're the man!  THanks for the translation, I think I had it right the first time, but wanted to be sure...  

And yeah, I'd rather have my NJB3 modded than my AD...

I have one of those Hosa ODL 276 coax>opt converters but haven't run it yet...

Terry

[F.O.Bean]

Ok, I'll attempt english, but it might turn out to be techno-babble, who knows.

The true SPDIF standard is -0.5volts for a logic 0 and +0.5volts for a logic 1.  Most gear doesn't really follow the SPDIF standard exactly (probably since the spdif standard is assinine).  So A/D gear will put out 0volts for a logic 0, and 2-5 volts for a logic 1.

Semiconductor chips (ICs) in general require "TTL" logic levels--transistor-to-transistor logic.  Theoretically, TTL levels are 0volts for a logic 0 and 5volts for a logic 1.  In practice, most ICs require somewhere 3.2 and 3.6 volts to recognize a signal as being high enough to be called a logic 1 (this is the 3.3volts Doug is talking about).

The trouble is the voltage level of the digital signal coming over the coax from the A/D converter might be high enough to meet or exceed the spdif standard, but it is not high enough to meet the TTL standard required by the ICs onboard the JB3.  This is the case, I believe, with my V3.  Apparently according to Doug, the Sony SBM1 puts out a TTL level signal on its 7-pin out (meaning, more than ~3.3volts for a logic 1, meaning alternatively that Sony does not conform to the spdif--Sony/Philips Digital InterFace--since they know the spdif standard is moronic).

So if you're using a SBM1, you could replace the optical jack with a 1/8" jack used for a coax input.  Tie this right to where the old optical went on the board (well, with the de-coupling capacitor Doug notes on the hot lead) and all would work fine.  But then when you upgrade from the SBM1 to a V3, it would not work since the V3 doesn't meet TTL levels.

You can get buffering chips that will take lower than TTL levels and boost them to TTL levels.  A good coax-to-optical converter design for example would do this.  I imagine this is what is done in say a Hosa coax>optical converter.  The crappy infamous "greybox" converter (which I bought and have lying around somewhere) does not use these buffering chips.  Which is why the greybox converter does not work with the V3.

Doug also says that rather that trying to install the buffering chips onboard the JB3 so it can take any coax output from an A/D converter, it might be easier to modify your A/D converter to be sure it sends out TTL level signals over its coax out.  The trouble with this approach is that if you change outboard A/D converters, your mod'd JB3 may stop taking the coax signal.

Bottom line to all this is that it just won't be too easy since the most flexible way will be to do all the changes on the JB3, but this will require adding circuitry to an existing surface-mount board.  I've thought numerous times of trying to attack it, but I don't want to mess with my JB3 and I haven't gotten motivated enough to buy a second JB3 that I can f*ck with and probably destroy.

-Todd

+T, you're the man!  THanks for the translation, I think I had it right the first time, but wanted to be sure...  

And yeah, I'd rather have my NJB3 modded than my AD...

I have one of those Hosa ODL 276 coax>opt converters but haven't run it yet...

Terry

terry, if you dont think youre gonna get around to using that hosa box, ill but it off of ya on thursday if ya have the wall wart

i was gonna call full compass otherwise!!!

im just gonna do transfers from dat>jb3>firewire

[twatts (pants are so over-rated...)]

terry, if you dont think youre gonna get around to using that hosa box, ill but it off of ya on thursday if ya have the wall wart

i was gonna call full compass otherwise!!!

im just gonna do transfers from dat>jb3>firewire

No, I'm going to keep it (thanks Jason Sobel), but may not really ever use it...  As for DAT>NJB3, yeah, that's what I do - I have an opt out on the DAT...

But I can see how you want to go DAP1>NJB3>FireWire...  You'll need a converter for that one right???  And too bad your D8 is broken, you could go 7-pin opt out to the NJB3...

Good luck with all that...  I wish the damned GMINI would get worked out, or we could get a hard input for the NJB3.  I just don't like opt in the field...

Terry

[F.O.Bean]

yep, dap1>odl-276>jb3>firewire>wavelab/samplitude(resampling/saving)>cdwave>flac

[Lee]

Ok, I'll attempt english, but it might turn out to be techno-babble, who knows.

The true SPDIF standard is -0.5volts for a logic 0 and +0.5volts for a logic 1.  Most gear doesn't really follow the SPDIF standard exactly (probably since the spdif standard is assinine).  So A/D gear will put out 0volts for a logic 0, and 2-5 volts for a logic 1.

Semiconductor chips (ICs) in general require "TTL" logic levels--transistor-to-transistor logic.  Theoretically, TTL levels are 0volts for a logic 0 and 5volts for a logic 1.  In practice, most ICs require somewhere 3.2 and 3.6 volts to recognize a signal as being high enough to be called a logic 1 (this is the 3.3volts Doug is talking about).

The trouble is the voltage level of the digital signal coming over the coax from the A/D converter might be high enough to meet or exceed the spdif standard, but it is not high enough to meet the TTL standard required by the ICs onboard the JB3.  This is the case, I believe, with my V3.  Apparently according to Doug, the Sony SBM1 puts out a TTL level signal on its 7-pin out (meaning, more than ~3.3volts for a logic 1, meaning alternatively that Sony does not conform to the spdif--Sony/Philips Digital InterFace--since they know the spdif standard is moronic).

So if you're using a SBM1, you could replace the optical jack with a 1/8" jack used for a coax input.  Tie this right to where the old optical went on the board (well, with the de-coupling capacitor Doug notes on the hot lead) and all would work fine.  But then when you upgrade from the SBM1 to a V3, it would not work since the V3 doesn't meet TTL levels.

You can get buffering chips that will take lower than TTL levels and boost them to TTL levels.  A good coax-to-optical converter design for example would do this.  I imagine this is what is done in say a Hosa coax>optical converter.  The crappy infamous "greybox" converter (which I bought and have lying around somewhere) does not use these buffering chips.  Which is why the greybox converter does not work with the V3.

Doug also says that rather that trying to install the buffering chips onboard the JB3 so it can take any coax output from an A/D converter, it might be easier to modify your A/D converter to be sure it sends out TTL level signals over its coax out.  The trouble with this approach is that if you change outboard A/D converters, your mod'd JB3 may stop taking the coax signal.

Bottom line to all this is that it just won't be too easy since the most flexible way will be to do all the changes on the JB3, but this will require adding circuitry to an existing surface-mount board.  I've thought numerous times of trying to attack it, but I don't want to mess with my JB3 and I haven't gotten motivated enough to buy a second JB3 that I can f*ck with and probably destroy.

-Todd

Thanks for the translation, Todd.  There's definately room inside the JB3 casing for extra circuitry... I'd almost be willing to donate my JB3 to the cause just to see if something could be worked out.  Here's hoping...

[John R]

Thanks for the translation, Todd.  There's definately room inside the JB3 casing for extra circuitry... I'd almost be willing to donate my JB3 to the cause just to see if something could be worked out.  Here's hoping...

not till after this summer

jr

[sickrick43]

I'd be willing to sacrifice line in for a standard TOSLINK socket, so at least you could get the cable to lock in place.

As far as the V3 goes, there's a jumper where you can select PRO or CONSUMER on the coax/AES1 output (Jumper ADJ7).  Page 6 of the manual describes this jumper but does say WHAT the difference between PRO and CONSUMER actually is.

Who's gonna be the first to call and ASK (and of course POST what the answer is).  Ask em when the OPTICAL OUT factory mod is gonna be ready to go, so we can quit these frikkin Hosa/C02 boxes.

Rick

[dklein]

As far as the difference b/w consumer and pro, here's a favorite site I always end up at http://www.epanorama.net/documents/audio/spdif.html

Go almost halfway down and look for Converting between AES/EBU and S/PDIF interfaces

[sickrick43]

We're not talking about an XLR/AES to COAX/SPDIF conversion here.  The manual doesn't describe the difference, but I suspect it's in the subcode header thats send (SCMS & ID's)

Rick

OK, I'LL call Grace in the morning and ask...

[Todd R]

As far as the V3 goes, there's a jumper where you can select PRO or CONSUMER on the coax/AES1 output (Jumper ADJ7).  Page 6 of the manual describes this jumper but does say WHAT the difference between PRO and CONSUMER actually is.

I'm not sure where you're going with this, but the pro/consumer setting on the V3 is not going to help with what I was describing above.  The standards for AES/pro standards and Coax/spdif standards deal with both subcode info and the electrical level of the signal.  Manufacturers of equipment regularly deviate from the electrical portion of the standard, as it ultimately isn't too important.  As I noted in my earlier post, oftentimes the signal level sent for a logic 1 is far above what is required by the spdif standard.

The V3's settings for consumer/pro change the subcode portion of the information being sent, but they do not change the voltage level of the signal.  The problem I noted above is that the V3 does not send a high enough voltage level out of its digital outputs to be read as a logic 1 by TTL ICs.  So the issue is not the subcode info, but the electrical side of the equation--the voltage level of digital 0 vs digital 1.  Changing the pro/consumer will not change this, so for instance, you won't get the V3 to work with the "greybox" converter no matter what the pro/consumer setting is on the V3.

I'm still anxiously awaiting Grace to unveil the optical out option to work with the JB3.  But still, I'd rather have a coax in on the JB3.  I try to be careful as I can, but just the other day as I was tearing down, I broke the little optical mini end of my optical cable.  I'd much prefer to be using coax cables.

-Todd

[dklein]

This thread was just crosslinked and it got me thinking...if you run a jumper wire from two of the 1/8" analog tabs over to the leads where the optical input connects, you could then use a standard 1/8" connection plug to tap into the JB3s s/pdif input electrically, bypassing the optical front end.

A mod like this would allow you to maintain everything - the analog in, electrical s/pdif and optical s/pdif.  Then, if there were any voltage level issues (depending on the a>d in use) you could deal with those outside the JB3 with an appropriate box.

Yes, you'd be jamming some electrical signal into the analog inputs at the same time but as long as you select optical in it would presumably be ignored.  If anyone thinks this might blow the analog in, then I suppose you could lift the analog tabs or perhaps choose the L+ and R+ connections as the s/pdif carrier.

Whaddya think?

[jk labs]

This thread was just crosslinked and it got me thinking...if you run a jumper wire from two of the 1/8" analog tabs over to the leads where the optical input connects, you could then use a standard 1/8" connection plug to tap into the JB3s s/pdif input electrically, bypassing the optical front end.

A mod like this would allow you to maintain everything - the analog in, electrical s/pdif and optical s/pdif.  Then, if there were any voltage level issues (depending on the a>d in use) you could deal with those outside the JB3 with an appropriate box.

Yes, you'd be jamming some electrical signal into the analog inputs at the same time but as long as you select optical in it would presumably be ignored.  If anyone thinks this might blow the analog in, then I suppose you could lift the analog tabs or perhaps choose the L+ and R+ connections as the s/pdif carrier.

Whaddya think?

Passing DC has some drawbacks. The S/PDIF signal is RF (MHz) so I suggest you try using 560pF or so caps instead of wire links. You'd need to experiment with the exact values of the caps. The aim would be to get sufficient RF signal level across the cap at 1 MHz and above in order to enable secure detection of the S/PDIF transitions. But at the same time small enough to effectively block alternative paths for the analog signal in it's search for ground. Probably you'll need more signal voltage on the feedline than the standard 0.5 Volts peak to peak specified for S/PDIF but thats a minor challenge here.   

I imagine the analog inputs have some overvoltage protection. If not they can be protected. If they are protected they will clip the signal received at some level related to the voltage on the power rail available. Anybody actually measured the logic voltages inside? Is it 3.3 or 2.5 Volts?

There is a ground issue here as well. Are the analog and digital grounds within the jb3 separated (as they ougth to be)?  One needs to think carefully about how to connect the coax shield to proper ground. Using L+ & R+ as you suggest needs to be evaluated. Connecting say R+ to a noisy digital ground could have some real ugly consequences wrt the analog capability (already suffering from noise according to user feedback.

So I think it is possible to add coax input capability to the jb3 analog input port without sacrificing analog performance.

However, based on a brief look at the pictures posted by Lee, it's not clear to me if you can just dump RF energy onto the output of the optical receiver. A photodiode f.ex. will not like that. Neither will a low impedance output buffer (it will kill the signal). Removing the optical in capability is an option. If you insist on keeping it you might be forced to adding a little bufferstage to protect the optical sensor. 

So I say go right ahead and keep us updated !!

Jon

[dklein]

So I say go right ahead and keep us updated !!

ha ha - I'm not hacking into my JB3 for a mod I don't need!