Schematics that you've learned something from

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

Rochey

Well-known member
White Market Member
Joined
Jul 2, 2004
Messages
2,697
Location
Dallas, TX - Formerly UK
Hello folks,

Thought I'd start a thread about commercial schematics that you may have seen that really taught you something.

I personally have spent hours pouring through some of the old Symmetrix Audio schematics, as well as some of the Rane schematics.

For example, I spent HOURS understanding the schematic used for the Genx6 clock distribution circuit, and felt that I'd really learnt some clever tricks.
http://www.symetrixaudio.com/kb/GENx6_sch.pdf
I think you learn a lot about a designer by looking at some of their work. The guy that did the Genx6 really comes across as an old school analog designer.
His (or her) design around the TLC2932 PLL device (Sheet 2) as well as the control/led circuitry (Sheet 4) really is a sight to behold.

Anyone else care to share a similar "ah-ha!" moment they got from reading another persons schematic/layout design?

Cheers

/R

BTW -- now that I'm reading this, I do realize how much of a Dork that I may appear to be.  ;)


 
I voraciously digested every schematic I could get my hands on for my first couple decades trying to design. It finally got to the point where the "ideas for design', published in the monthly rags were just rehashes of old ideas, but still interesting to find the "secret" mistake published in every one.  ;D  In fact back before the tight integration between schematic and PCB CAD, more schematics had errors than didn't. Always kept it interesting, and worthwhile to really understand the circuits.

The William's book, "Analog Circuit Design" shares several well respected engineer's, personal inspirations or Ah Ha moments.

JR

 
JohnRoberts said:
I voraciously digested every schematic I could get my hands on for my first couple decades trying to design.

+1.

Both good and bad designs are interesting in their own ways; one more thing I liked about H&H's Art of Electronics is how all chapters are concluded with a collection of good ideas and buggy circuits.

Rochey said:
For example, I spent HOURS understanding the schematic used for the Genx6 clock distribution circuit, and felt that I'd really learnt some clever tricks.
http://www.symetrixaudio.com/kb/GENx6_sch.pdf
I think you learn a lot about a designer by looking at some of their work. The guy that did the Genx6 really comes across as an old school analog designer.
His (or her) design around the TLC2932 PLL device (Sheet 2) as well as the control/led circuitry (Sheet 4) really is a sight to behold.

What did you like in particular? The diode logic? The supply shifting for the TLC2932 depending on whether narrow or wide operation was selected?

(This old RF grump is a bit disappointed to see that there are quite a few unnecessary jitter injection vectors, especially in the selection/distribution matrix on page 3).

JDB.
 
yes, the supply shifting on the PLL device was pretty cool.
General use of Diodes and discrete transistors for logic is cool too.

By the way, you mean issues with the 74hct393 on page 3, or the buffer schmitt triggers?

your right, there does seem to be a lot of extra logic that may not be required. (in my opinion...)

tell me more about the "jitter injection vectors" -
 
Rochey said:
tell me more about the "jitter injection vectors" -

First, much of this is polishing the proverbial turd. Once you're clocking off a '2932, you're not going to get stellar jitter performance, and all other factors may or may not disappear into the noise floor.

Having said that, here's the five minute breakdown.

The big, avoidable injection vector is in U14. This ripple counter will keep on dividing and generating wideband noise, even when it's not being used (as in superclock out). That will effect even the XO modes. This could have been avoided by driving the CLR input of U14A with another gate or some diode logic, so that the counter is enabled only when used.

To a lesser extent the same goes for U17. There is non-negligible crosstalk between gates in the same package, and different gates are being driven at the same time for some operating modes. This is handled better for the mux U19A: as long as you don't plug in AES when you don't use it, only one input is driven with a clock at any time.

Same for U6: gates in the same package are used both for WC in and WC out on the PLL. I can see that there are economic considerations for doing it this way, but you will get leakage. Also, schmitt triggers are more noisy than regular gates.

Back to U14: a ripple counter is a good way to get more jitter in your clock. A synchronous counter would have been better; several are available at similar cost.

Output drivers: AC(T) logic is notoriously ringy/noisy, although that might be dampened a bit by the output resistors. Worse, they're much faster than required, which is bad for EMI (and ultimately jitter) Simple cheap HC logic would have been better, although possibly at the expense of needing more gates to get the same drive strength.

The WC input termination (R16, R12) is ... odd, effectively putting a DC voltage on the input when activated.

JDB.
 
Once I bought a Sony mxp2900 mixer, with complete manual with ALL schematics.
Just to get it up and running I had to read those schematics after which I became interested and joined tech talk..the previous forum.
 
> I voraciously digested every schematic I could get my hands on for my first couple decades trying to design.

+1
 
It changes all along the learning curve. I remember how I was drooling on a crystal set schemo, then a two-transistor radio, when an OC45 cost a whole month allowance, and the Truvox tape recorder, also the first of a 100W solid-state amp. Later, I've always been impressed by almost everything dbx made (the 174 gate is a perfect example of over-engineering disease ;) ). And I concur with Samuel; the AP gear is the epitome of thoroughness. In almost any piece of gear, there is something to learn or to despise :)
 
A good and hard school for me was all 1970-1980's Japanese service manuals for high-end HiFi gears.

Like Accuphase, Denon, Harman-Kardon, Luxman, Marantz, Kenwood, Nakamichi Onkyo, Pioneer, Sansui, Sanyo, Sony, Stax, Yamaha and many other.

When they all in this time design discrete op-amps, with very complex topolygy of cascode, current mirror, constant current source and other interesting things.

I have lot of thanks to all them.

But of course all pro-studio mixing concoles manuals was also a very good school, to get knowledge of all new type of paramitric equalizers, mike-amps, mix-busses and much others.

--Bo
 
Samuel,

Yes, but it is a lot of work to scan, many time very big size sheets of schematics.

But start to download a lot of nice schemaics here:
http://www.hifiengine.com/manuals.shtml
http://www.audiocircuit.com/index.php
http://www.acousticpsychos.com/Files/
http://www.audio-circuit.dk/Schematics.html
http://www.abeltronics.co.uk/diagrams.php

I think you have a lot of time here, to print out all interesting manuals/schematics.

--Bo
 
Back
Top