DIY SQ Decoders?

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Thanks.

For the resistors, I'm using SMD, but for the caps it's going to be a mix of SMD and through-hole. Odd value resistors are easy to get, but getting film-type caps in some of these values and tolerances is tricky, and taking what I can find in whatever package I can find makes it easier and cheaper.

I don't want ceramics in the audio path... especially in the decoder, so smaller than 1206 is pretty unlikely anyway.

I had been using 0402 resistors, though, but they are pretty small... I will see whether the values I need are available in larger packages.

Edit: Just checked, and it looks they are available in 1206 1/4W for the same price, so I have changed them out in the schematic.
 
I built an SQ decoder based on the Motorola 1312 at Uni in 1977 with ceramic & film capacitors. Ceramics aren't as bad as they used to be, but I would avoid any leaded or SMD capacitors with Y5V dielectrics as they are very voltage dependent, and I'd be a bit careful with X5R/X7R as well at larger voltage swings

Thanks.

For the resistors, I'm using SMD, but for the caps it's going to be a mix of SMD and through-hole. Odd value resistors are easy to get, but getting film-type caps in some of these values and tolerances is tricky, and taking what I can find in whatever package I can find makes it easier and cheaper.

I don't want ceramics in the audio path... especially in the decoder, so smaller than 1206 is pretty unlikely anyway.

I had been using 0402 resistors, though, but they are pretty small... I will see whether the values I need are available in larger packages.

Edit: Just checked, and it looks they are available in 1206 1/4W for the same price, so I have changed them out in the schematic.
 
The ceramics I was finding are C0G multi-layer ceramics. If those would be suitable, they would definitely bring the cost down.

The Kemet data sheet claims no piezeoelectric noise, which is one of the concerns I had about ceramic. They also say C0G is "stable", and no capacitance variance with respect to time or voltage.
 
C0G or NPO capacitors are very good, though getting high values can be difficult as they get physically very large. Piezoelectric noise is only a problem when the PCBs flex as the effect is like a microphone, but you need quite a bit of movement. So placing sufficient fixing holes (including one in the middle of the PCB) to give the board rigidity, and screwing down the PCB onto a metal plate will be sufficient. Put the unit you're making into a metal box to electrically screen it as that is the most likely form of noise/interference you'll get.

The ceramics I was finding are C0G multi-layer ceramics. If those would be suitable, they would definitely bring the cost down.

The Kemet data sheet claims no piezeoelectric noise, which is one of the concerns I had about ceramic. They also say C0G is "stable", and no capacitance variance with respect to time or voltage.
 
I’ve stumbled across a couple of sets of the Motorola ICs that I stashed away some years ago. I feel I really ought to try and use them to make one of the CBS circuits (I don’t really need to as I’ve got an Involve SM for everyday use, but I’d like to see how it would compare to my venerable old Sony SQD-2020). The only thing that’s putting me off is creating the PCB – It’s been a long time since I etched my own boards! I don’t have the necessary CAD software (or the inclination to use it) to create the sort of files that prototype manufacturers demand these days to etch a board for you, so it will have to be the rather labourious process of creating photo resist artwork by hand – unless anyone knows of any dusty old blank boards for such circuits lying around somewhere of course?!
 
I’ve stumbled across a couple of sets of the Motorola ICs that I stashed away some years ago. I feel I really ought to try and use them to make one of the CBS circuits (I don’t really need to as I’ve got an Involve SM for everyday use, but I’d like to see how it would compare to my venerable old Sony SQD-2020). The only thing that’s putting me off is creating the PCB – It’s been a long time since I etched my own boards! I don’t have the necessary CAD software (or the inclination to use it) to create the sort of files that prototype manufacturers demand these days to etch a board for you, so it will have to be the rather labourious process of creating photo resist artwork by hand – unless anyone knows of any dusty old blank boards for such circuits lying around somewhere of course?!

Most of the necessary CAD and routing software, like Eagle CAD, is free for two-layer boards. Should be fine for this. The way people do things today is use that free software to create the layout and then send it directly to a small batch board company that can usually make them really cheaply in China and mail them directly to you. Through-plated holes and vias, silk screen, heavy copper, the whole nine yards. Most of these companies can even pick and place the parts you specify from a Mouser or Digi-Key catalog.
 
Hi SQLGuy - It's not a question of the expense or otherwise of the software, rather it's an issue of old dog and new tricks! I see you were using it to generate an L3 pcb - how did that go?
 
I would be interested in that. The L1, L2, and L3 circuits I was referring to are from this document: http://www.ka-electronics.com/images/pdf/CBS_Motorola_SQ_Decoders.pdf

Looks like the main advantages of L3 are lower distortion and lower noise by using the discrete matrix instead of the MC1312.

The main improvement is the accuracy of the all-pass, phase shift network. The network normally used with the MC1312 is four pole. My Audionics 106A (Basic SQ decoder board) had a six pole network and was noticeably better. The Fosgate Tate decoder also has a six pole network. The Audionics S&IC uses an eight pole network.
Back in the day I bought an SQ decoder module form Heathkit that utilized the Motorola chips. That design utilized the variable blend circuit on the rear, similar to the circuits in your document. Loving the performance of the Audionics board I piggy backed it on the Heathkit board substituting it for the MC1312 based decoder. I used that for several years before getting my S&IC. As I was always fiddling with it I ended up having to replace the MC1315 chip a couple of times, then suddenly they were no longer available. Now Chinese sellers have all of them them available cheaply! Unfortunately I can no longer find that decoder board, I suspect that it was thrown out (not by me) at some point.
The circuit is not that complex, it could be built up on a perfboard, still PC boards are nice and would construction a breeze.
 
Hi SQLGuy - It's not a question of the expense or otherwise of the software, rather it's an issue of old dog and new tricks! I see you were using it to generate an L3 pcb - how did that go?

I got distracted by buying a Lafayette SQ-W that needed a bit of work. I repaired it and have been using that. Current distractions are more around needing to replace all the electrolytic caps in the SQ-W and needing to recalibrate it. It works well enough, but I think it will probably benefit from such an overhaul. I have the parts, but haven't made the time to do the work.

Mostly I've been focusing on techniques for digitizing output from the SQ-W and eventually getting that content to DVD-A.
 
I got distracted by buying a Lafayette SQ-W that needed a bit of work. I repaired it and have been using that. Current distractions are more around needing to replace all the electrolytic caps in the SQ-W and needing to recalibrate it. It works well enough, but I think it will probably benefit from such an overhaul. I have the parts, but haven't made the time to do the work.

Mostly I've been focusing on techniques for digitizing output from the SQ-W and eventually getting that content to DVD-A.

Good luck with your projects - I understand the problem of distractions!
 
Hi there, I have made a few decoders using the Motorola SQ 3-chip sets they are pretty easy to cobble together either with veroboard of basic pcb layouts.

I have about 6 sets of these chips for sale - US$25 for complete set of 3 units.

Anyone interested then please
PM me.

thanks

Steve in New Zealand.
 
Hi there, I have made a few decoders using the Motorola SQ 3-chip sets they are pretty easy to cobble together either with veroboard of basic pcb layouts.
I have about 6 sets of these chips for sale - US$25 for complete set of 3 units.
Steve in New Zealand.

Hi, please foto ... thx
 
Back
Top