Musical Fidelity X-PSU DIY

[NeutralMAN]

Hi all,

I am in the process of "attempting" to build a X-psu for the x-10 v3. I know the psu type is a 12-0-12. The standard wallwart plugpack is rated at 500ma. I have been able to get my hands on a toroidal transformer rated at 160VA with two 12v secondaries (over 6 amps :E ).

Would anyone have some advise before I fry the insides of the x-10 with the DIY X-PSU?? The questions I have include:

- Is the 160 VA transformer to large?
- Will this large transformer cause problems in the regulation circuits?
- When running the secondries in series, where does the 0 volts wire connect to? I'm assuming the centre wire of the x-10 plug?

It would be good to start a thread on how to build your own DIY X-PSU. I have found instructions for the older v2 psu. The older one operated at 12v, and the newer v3 operates at 24v.

I'm assuming alot of members will benefit from a DIY X-PSU.

Thank you,

NeutralMAN

[Taito]

Hi, although I don't have an X-10, I have considered buying one, and have given designing a suitable power supply some consideration.

To work out which wire of the x-10 plug accepts which potential (-12,0,+12), why not hook it up to its wall plug and use a multimeter to measure the voltages present on the terminal for the X-PSU? May or may not provide you with the information you need. If you go through with this, you need to make sure of what voltage goes where, or you run the risk of cooking your X-10.

Since you are using regulator IC's (I assume), check the specs and current limitations of these. While the 160VA rating is a maximum rating, it is overkill if the regulators cannot take advantage of the transformer's maximum current delivery.

Otherwise, a convential supply with a decent amount of capacitance and high quality regulators should provide you with good, stiff supply rails. If you are suffering from noise, it wouldn't hurt to run small capacitances between the pins of your regulators (input to ground, output to ground). You should end up with only a very small amount of ripple on your supply rails.

Hope this helps, Ben

[TacoD]

also check http://www.rock-grotto.co.uk/x-psu.htm

Taco

[NeutralMAN]

Thanks All,

I have seen the rock grotto project before, but his is for the x-psu V1-2, not the V3.(12v compared to 24v)

I was under the impression, that the regulation and smoothing occurred inside the x-10 v3. This occurred with the v2, bu not sure about the v3.

Taito,

I did try the multimetre test. When using the centre pin as neutral, the left right pins produced 12 volts each (13 actually). When wiring up the toroidal with series secondaries, I'm assuming that the point were the secondaries are connected, is used as the neutral wire (centre one described above). Am I correct?

NeutralMAN

[Taito]

Yes, based on what you have said, using the centre-tap of the transformer - the point where the secondaries are connected will be the 'ground' reference. Whether or not you choose to attach this point to mains earth or not is up to you. Safety-wise it is best to earth you chassis, but you may find that doing so injects some noise.

-Ben

[NeutralMAN]

Taito,

The usual MF wallwart is not grounded. Now since the wallwart is not even grounded, I'm assuming MF doesn't earth this grounding point. (DUH!!)

I was going to use an alluminium case with shielded power cables and earth both. But I wasn't going to connect the "0 volt ground to earth". Should I be?? , even though the case and power cable will be earthed??

Now I'm not the smartest around electronic theory, but won't the current/power drain from the "ground" to earth continually??

Finally, I do not know much about adding a capacitor across the input termnials to eliminate noise. What sort of values do I need for the Australian standard of 240v 50hz?? It goes against every saftey instinct, since I know it's not good to connect +VE and -VE power terminals. Can you elaboratre further?

NeutralMAN

[Taito]

In this application, it is probably simpler not to connect your reference ground to mains earth (I would recommend trying it though - if there isn't a discernable drop in performance, it is safer). The way doing this works, is when you have a DC supply with +12V,0V,-12V, these voltages are really potential differences. So, +12V is 12 volts higher than 0V. But what is 0V? If it is tied to mains earth, then 0V and mains earth are at the same potential. +12V is still 12V higher than this and -12V is still 12V lower than this. Your components wont care what the reference is, just that they have a potential difference across which to operate.

Adding a capacitance between the input pin of each regulator and the ground pin and also between the output pin and the ground pin is a way of reducing HF noise on the supply rails.

For theorectically ideal components, a resistance is a purely real impedance that presents the same impedance across all frequencies. A capacicor is... scap that - probably more complcated than it needs to be.

A capacitor's impedance is dependent on frequency. It will block DC and pass high frequencies. Therefore there will be no DC power flow from 12V to ground, but HF noise will be allowed to flow from your rail to ground, resulting in PS rails with less noise. This is good for performance.

Hope this clears thing up, Ben

[Taito]

Sorry, just saw the request for capacitor values. Without doing any calculations at the moment, try, say 100nF.

[NeutralMAN]

Thanks for the input,

I'm just in the process of buying a dremel tools, that will allow me to cut proper holes in the 3mm thick alluminium box.

Will let everone know about the progress of the X-pus V3 clone :T

NeutralMAN

[Taito]

Well, good luck. I'll be very interested to know how it turns out! If you have any questions, don't hesitate to ask.

-Ben

[Occam]

Neutral -

Why use a toroidal transformer at all?
Unless your toroid has a shield between the primary and secondary, the ONLY benefit that a toroidal transformer provides in a minimal external magnetic field. As you're housing the ps in, I assume, and external box, that benefit should be moot.

A split bobbin EI core transformer will provide the following benefits -
1.lower bandwidth (all you want to pass is the line frequency, 50/60hz
2.far less capacitive coupling, which will pass far less hf noise and crap which permeates most of our mains.
FWIW

[NeutralMAN]

Is there any place the get the shielded toroidal in Australia? I already have the a toroidal at the moment.

Doesn't the origional plug pack have a IE tranny in it?? What is a split bobbin?? is it different than the plug pack?

I was under the impression that the toroidal is better than the IE tranny.

I wanted to keep the first go of the X-psu as simply as possible.

Thank you,

NeutralMAN

[Occam]

Neutral,

"I was under the impression that the toroidal is better than the IE tranny."
Thank goodness technology is not democratic.

Toroids with shielding between primary and secondary are very expensive, generally only found in medical isolation transformers or high quality mains balancing conditioners. No doubt, Antrim would be more than happy to provide you a custom shielded toriodal. They supply Maplin [or at least used to] their non-shielded toroidal transformers.

Adequate split bobbin transformers may be sourced in Australia at RS Components. A 20va dual 12v winding unit is stock# 805-063 @ AUS$ 16.10 or the 50va 805-136 AUS$ 29.70. I personally prefer over specifying transformers, but either should do.

http://www.rsaustralia.com/

Then again, you are quite free to 'piss into the wind'

FWIW

[NeutralMAN]

Occam,

What is the difference between the normal IE tranny and the split bobbin?? Sorry but I am not a expert when it come to electronics, so some additional informtion would be greatly appreciated.

What difference with performance would you expect between the toroidal and split bobbin??

The only reason I stuck with the toroidal, is that it was easy to get, and MF used it in their PSU. Now if the split bobbin is way better, ,then I might go with it.

Also I have noticed shielding does help my components immensly, and would like to investigate the split bobbin.

Thank you,

NeutralMAN

[NeutralMAN]

I also have now read that the split bobbin tranny, is still better than the shielded toroidal. Is this correct?

NeutralMAN

[Occam]

Neutral,

As to whether a split bobbin EI core transformer is superior to a toroid with an interwinding shield, I'd say yes, all things being equal. But they're not equal, as you're comparing apples to donuts.
For minimizing the capacitive coupling between the primary and sencondary, which limits the pass-thru of noise, yes because you simply physically separate the windings. As the EI split bobbin mains transformers you typically source in Australia are also sold in Europe, they are CE compliant, and the 'creepage' requirements of compliance force that physical separation of the bobbin. Because of this, almost all EI transformers are split bobbin rather than the previous construction where the primary and secondary are wound over each other, which had large amounts of capacitive coupling, like the typical toroidal transformer. Similary, both can address this coupling issue with a grounded shield between the windings, an extra, high tolerance manufacturing step, which is the reason for the increased cost.
For limiting the bandwidth of the transformer, again yes, as a mains toriod actually makes a resonable audio transformer.
But for containing the magnetic field, which can induce hum, the strong advantage goes to toroids. But in your situation, where you intend to house the transformer in a chassis separate from the actual audio macinations, the first 2 are of more concern than the latter. I've great admiration for anyone who can build a component with an EI core power transformer in the same box as low level active electronics and not have hum problems.

[Occam]

Someone else's x-psu journey -

http://www.rock-grotto.co.uk/x-psu.htm

read to the bottom....
FWIW

[NeutralMAN]

Hi All,

Sorry for the delay in my post, but have not had enough time to finish my PSU.

I finally got around to completing the unit.

From initial results, it really seems to increase the soundstage and the bass responce.

but...

Not sure on what value of fuse to use. I was using a .250 A and a .5 A fuse. The only problem is that is seems to blow most often. The secondaries are rated at 6+ amps.

Would someone know what the best value would be?
Would I need a fuse rated for over the max amperage for the secondaries?

I have read the Rock grotto site, and they use a 1 amp fuse. The psu they built was V2 (therefore non balanced 12V). The one for the V3 is a balanced 24v. Therefore, would I need at least a 2A?

Your assistance would be greatly appreciated.

Thank you,

NeutralMAN.

[cjd]

160VA trafo is going to blow those fuses only at startup (inrush current is high). A 1.5A should do the trick. My generic "rule" is to divide the VA by 100 to get an approximate value that will survive. On the 330VA I'm using in a project, 3A survives, 2A doesn't. Slow blow fuses too, if you aren't already.

Also, if you do wish to connect your 0V to earth and keep ground-loop induced issues at bay, use a 20-100ohm resistor to make the connection. In theory you should need not much wattage capacity, but...

C

[NeutralMAN]

Thank you cjd,

I'll give that a go. I have noticed that if I plug in a medical grade(??) power filter, the fuse doesn't blow that much. I'm assuming that the filter would be limiting the inrush current.

I just want to make sure that the 1.5 A fuse is enough to catch problems so as not to blow the X-10 V3.

In addition, I'm waiting for a lend of a camera. I will post some pictures of my DIY PSU when I get a chance.

Thank you,

NeutralMAN

[cjd]

A short should trip a 1.5A. Heck, I'll be up to 6A (may try 5A) on the amp I'm working on currently.

The other thing you can do is look into a slow-startup circuit. Usually a few hefty resistors, a cap, some diodes, a relay... I think. Don't remember all the pieces. Many people report that such a circuit actually reduces any transformer hum they had as well.

C

[NeutralMAN]

Thanks cjd,

I'll keep that in mind for future upgrades. I might be doing the slow-start and the capacitor filtering on the next iteration of the x-psu.

NeutralMAN

[cjd]

Technically, I'm working on a PSU for an X-amp too.

Have fun!

[NeutralMAN]

cjd,

What type of amp are you working on? x-80, x-150? and how is it going?

NeutralMAN

[cjd]

X-GC Flexible, but the main one should hit ~200WPC, double that peak. Driving it with a Twisted Pear Audio X-BOSOZ. I have a thread somewhere here on it.

Boards are now all assembled waiting for me to assemble IEC>fuse>switch>trafo and connect them up. Will be a few weeks til I can get back to it due to a mishap today while trying to make a b-day gift for my wife. :|

C

[Dennis H]

a mishap today while trying to make a b-day gift for my wife.

Yikes, sorry to hear that! Do you still have all your fingers?

[NeutralMAN]

That's not good to hear CJB, hope all is well.

NeutralMAN

[cjd]

Yikes, sorry to hear that! Do you still have all your fingers?

Yup. Damage seems to have been limited to soft tissue - tendons and bones are good. Wood sure flies off a table saw fast though! I have to see a hand surgeon to see how well they can close things up to minize scarring and hopefully not hinder my violin playing.

Not allowed to work either. Something about pain meds and driving not mixing. . . Typing is interesting.

C

[Paul H]

Sorry to hear about that Chris - hope you heal quickly and well.

Paul