advertisement


Best CD clock upgrade?

I think the coypright needs to read Tony L, or is PFM still a legal entity?
Click to expand...

Well, it exists as we're using it, I was going to reference the 'site' and the 'site owner' from which anyone can find Tony's contact details.

To be honest anyone wanting to sneak it into a commercial item is likely to do so (most won't as it's not cheap enough ;) ), my main concern is unauthorised re-selling to the DIY market or similar areas.

The main concern is to make people think twice.

Andy.
 
All good stuff this - seems there's always something new to learn!

For PCB manufacture, as dodgy looking as they might be, I've had good success with these:
http://www.goldphoenixpcb.biz/

Bought quite a lot of prototype runs from them. Cheapest was 100sq.in. for about £53 delivered complete. 2 layer PTH, 2 resist & 1 ident. Delivered typically in 10 days or less.
 
Hi Guys,

I've created a PDF of the board layout, I'd be very grateful if someone (maybe more than one) could look it over and check the schematic and the board.

I've added an HF bypass at the input using a small SMD, I don't think the cct will need too much more than this, although it's somewhat dependant on the length of power cabling to it. I've also added a single mounting hole in the only area available, without increasing the board size. Since the clock wants to be close to the circuits it feeds, in many case with a board of this size double-sided sticky tape would be sufficient to hold it.

There's only 4 easy surface mount components, so it should be real easy for most to build, if we're all happy with it, I'll generate CAM files and we can put it out to tender!

There's a single error at present on the XO module - the ident (XTAL1) is on all layers, not just the to silk, I'll correct that tomorrow as I forgot to bring the lib files home.

Any questions don't be afraid to ask, better we discover problems now, than later!

Andy.

The file is here in PDF form.

XO_Clock.gif
 
Wow, more addictive than Neighbours :)

Thank you for sharing that knowledge Andy, very informative and generous. I'll put the vero away now ;)

I've printed it off and am checking. I can confirm that the XO I have has 7.5mm pin spacing,

My Farnell order is currently on hold.

Cheers,
Eric.
 
Thanks Eric!

My Farnell order is currently on hold.
Click to expand...

It's worth checking suitable parts are available(!), I've not put great thought into the dimensions of parts, like L1 for example - currently it's resistor sized on the basis there's a few small chokes out there suitable for this, but don't take my word for it ;)

Andy.
 
You can get the Tent XO for £29 (inclusive of VAT and next-day delivery!) from Chevin audio (see the very bottom of the page) and it even comes with a choke that should go just nicely in a resistor-sized space. Perfect.

Great stuff Andy,
Car
 
Mine came with an unmarked, "resistor-sized" component that I have always logically assumed was the inductor as the web site says it comes with an "additional RF choke ("bead")".

Artwork looks good, just need to check pin-outs in the morning.

Edit - Pin outs look good. Now on to components sizes, might take a bit longer.
 
Is this beauty going to work in any CDP (mine is a Freq of 16.344)?

If so, put me down on the order list....

Jo
 
It will work for any frequency that Guido sells, which includes 16Mhz :)

I want some too, one for a reference in a phase noise measurement system, one to try in my CDP!

Andy.
 
Andy,
Just double checking a few things.
1] Is U3 correctly orientated, flat on other side?
2] C6 to be labelled as “C6”.

Cheers,
 
This "flea power" is wonderful work and a great learning experience.
What would be the ideal spec for L1?
I have been using Panasonic P9818 for these types of applications with success.
What does Guido ship with the XO and what does Audiocom use in SuperClock2?
Also, it would be nice to have option (or even sole possibility) to use AD797 in SOIC: this way things could be pushed even closer together, and perhaps another mounting hole could be added.
In my experience, when positioning these clock modules close to the chip they need to power, it is best to have a variety of holes to mount them to a pcb, since probably only, say, one out of four will line up above a section of the motherboard without pcb traces, to allow for drilling. You can probably tell that the double-sided sticky tape idea makes me nervous. :D
Andy, ignore my PM to you about hacking the SC2: you already answered fully on this forum, even if I would still like to see what is under the "blob". ;)
 
Eric,

U3 should be fitted topside, as a conventional leaded component. The symbol looks correct to me: -

http://www.semiconductors.philips.com/pip/BC547.html

Do you think it's incorrect at present?

James,

In many ways the easiest method to mount a clock such as this is to solder it to the ground plane of the CD player (most have one), using maybe 4 short wires at each corner. Might be worth trying to squeeze in some pads at each corner for that - I usually just solder something to the ground plane - I often use some thin braid (desolder braid works well!), as it acts as a bit of a compliant spring and gives some mechanical decoupling to the clock module and is also low impedance - crystals are inherently electro-mechanical so a bit of compliance is a good thing. I might give that some more thought!

I hate sticky tape too, so maybe the soldered down idea is good - more holes would increase the board size, I think it's worth keeping that down, so many are too big to implement well otherwise. SMD for it all would be perfect, but it does reduce it's DIY-appeal. Engineering is always a set of compromises :)

Andy.
 
What would be the ideal spec for L1
Click to expand...

It's not too critical but in an engineering sense it needs to do the following things: -

1. It needs to provide sufficient increase in impedance as frequency rises, to prevent the decoupling cap on the clock module from reducing phase margin in the AD797 regulator section.

2. It needs to provide some loss (i.e. low-ish Q) to prevent the LC combination of the choke and the clock decoupling capacitor forming a resonant circuit. This can be acheived by using suitable ferrites (which appear elecrically as if a resistor is placed across the choke, to damp it) or using a high-Q choke and damping it with a resistor. If the clock module comes with a choke, I'd trust Guido to have done his homework!

3. It ideally should provide an even impedance looking back from the clock to the regulator section - the AD797's output impedance will rise as the open loop gain falls off and the local decoupling at the clock output becomes ineffective w.r.t. the clock, so the local bypass 'takes over' and results in a steady reduction in supply impedance, up to the point the clock decoupling becomes reactive.

I think that just about covers it ;)

In reality, it's easier to determine by experimentation and isn't likely to be so critical in a digital application, providing things are well behaved and stable, which is easiest to see by probing with a 'scope (which I'll do once we have some boards).

I need an excuse to play with my new toy (gratuitous geek pic attached)...

boystoys.jpg


Andy.
 
Great job Andy, the layout looks fine to me as does the linked BC547 pinout

Re: L1 value - it works fine with everything from Guido's ferrite bead up to a handwound thing of about 5uH. The SMT decoupling at the XO module makes a much bigger difference to controlling noise - and ensuring stability. I've even used a 6.8uF SMT tantalum for the latter part without issue.

I've retested using an NE5534. It's just fine (leave out the 47pF cap!) but because there are larger input currents, the DC precision suffers a little - ie more current flowing through the 10K filter resistor means there's a slight voltage offset. We're only talking a few mV though, not enough to need to alter any circuit values.

Re mounting: I've deadbugged these things together on double-sided PCB material, and usually use a single copper wire soldered on the underside to the CD player ground plane as a leaf spring - it's electrically isolated and bounces well too.
 
Martin,

You've just made me think, for a connection to the player, the 0V connection is best made at the clock output, we would create a loop if we connect the input 0V too.

It's worth keeping as a test point, but maybe I should re-locate the clock outputs along the RH edge (or maybe front edge?) of the board, to make connection easier - the primary DAC - clock wires should be short and could act as a leaf spring, all one would then need is suitable insulation under the PCB to prevent shorting. The secondary feed is then made with a coax or twisted pair with 0V terminated at the sending end only.

What do you think?

Andy.
 
martin clark said:
I've retested using an NE5534. It's just fine (leave out the 47pF cap!)
Click to expand...

I've used the NE5532 in my current implementation (as per the original recommendation on the second page of this thread), but folks should be aware that the pin-outs are different... you can't plug an NE5532 into the above circuit and expect it to work, as I found out the hard way after delving into the datasheets! The correct one to use, as Martin points out here, is the NE5534.

Edit: but of course, if you're going to the trouble of building this, an AD797 is the best option, although far more expensive.
 
maybe I should re-locate the clock outputs along the RH edge (or maybe front edge?) of the board, to make connection easier
Click to expand...
That sounds good, you could just leave a strip of tinned ground plane at RHS next to the clock to terminate the coax shield / ground strap to. 10Mohm leaded resistors form quite handy and springy standoffs if needed!

The 0v point on the left can be left labelled for use with an isolated supply e.g. separate transformer.
 
You must log in or register to reply here.


advertisement


Back
Top