The Elektor Forum will close. See also this link. From Friday March 15st it is no longer possible to log in to the forum. However, the content of the forum will remain visible until the end of March. As of April 1st the forum will definitely go off the air.

graham slee gram amp2 need power supply circuit

HiFi, LoFi, Tubes, Transistors, ICs, Commercial, DIY

Postby dewisant » Fri May 10, 2013 12:00 am

is there a design out there for a 24 volt power supply for my 'special edition' gram amp 2 - cannot seem to get a second hand one and the cheapest I can find is around the £150 mark
dewisant
 
Posts: 3
Joined: Thu Jan 02, 2014 10:37 am

Postby thijsbeckers » Mon May 13, 2013 12:00 am

I think you need to provide some more details on the specs. Like how many amps, minimum ripple voltage, buffer capacitor size, etc.
thijsbeckers
 
Posts: 1134
Joined: Thu Jan 02, 2014 10:37 am

Postby dewisant » Mon May 13, 2013 12:00 am

hope this is what you are asking for
Graham Slee PSU1 Power Supply



The PSU1 is the unipolar linear audio power supply supplied with our flagship phono stage preamps and headphone amplifiers.

Safety certified to IEC standard 60950 - click here to download CB certificate (PDF file).

The PSU1 linear audio power supply has a stabilized output voltage of +24VDC as well as a nominal 30V DC, at a down-rated 200mA. It is available with detachable DC cords fitted with plugs to fit 2.5mm (standard issue) or 2.1mm DC socket pins, and with detachable mains cords to suit UK, European, Australian or USA outlets (specify at time of order).

Two mains voltages are available: 115VAC and 230VAC. They have sufficient latitude to accommodate voltages from 90 - 130VAC and 180 - 260VAC respectively, when powering Graham Slee products.

The PSU1 is designed to be a remote power supply placed well away from sensitive equipment like turntables and cartridges which are susceptible to pick up of stray magnetic fields. For maximum isolation it is best sited off the hi-fi rack.

It uses a specially manufactured mains transformer made using top quality transformer steel thus consuming minimal excitation (core magnetisation) current whilst optimising transformer regulation. A benefit is that it keeps transformer buzz to an absolute minimum which is only audible at very close quarters.

Transformer secondary, rectifier and reservoir capacitor interconnections are kept to the absolute minimum length minimising charging current losses and interactions.

The output is regulated at a stable +24VDC suitable to power all existing Graham Slee audio products featured on this website (In April 2010 an unregulated 30VDC (nom.) output will also be provided for users wishing to use their own regulator).

The PSU1 can be used to upgrade the sound quality - sound-staging, more vivid tonal colour and bass detailing - of Graham Slee phono stage preamps and headphone amplifiers supplied with switch-mode or linear plug-top power supplies. Upgrading to the PSU1 has the effect of taking your Graham Slee phono stage preamp or headphone amplifier to a higher level of musical performance.

Output cable accusations
Once in a while we are berated for seemingly not knowing our job regarding the DC output cable of our PSU1 power supply. Such criticism makes its way around internet forums, probably peddled by our competitors, and before you know it many have been convinced by outright lies. Below we prove the critics wrong using real world physics, math and manufacturers data.

Shielding is required for signal cables where there is some source impedance (driving impedance). Obviously there must be some source impedance because a signal cannot come from, or propagate around a short circuit! (tip: try proving it using Ohms law). Because there is source impedance radio frequencies can be induced onto signal cables and therefore signal cables should be shielded.

The DC output of the PSU1 does not supply a signal (although Kelvin currents return to it which will be covered elsewhere) - it produces DC (direct current). It is a virtual a short circuit to signal currents including radio frequencies - it has virtually zero source impedance - similar to the output of a power amplifier, it therefore does not need to be shielded. Here's why:

Take a look at the capacitor impedance curves below. If you scale over the left-hand V notch to between 3.3uf and 10uf you will get the impedance curve of the output capacitor of the PSU1. The bottom of the notch will be at 800kHz (middle of MW AM radio reception) and the impedance will be approximately 0.012 Ohms. Because all capacitors exhibit inductance, the impedance begins to rise above 800kHz, so lets approximate the impedance at 4MHz (lower end of SW AM radio reception). It is between 1 and 2 Ohms.
[URL=http://s21.photobucket.com/user/dewi123sant/media/capacitor-impedance-curves.jpg.html][IMG]http://i21.photobucket.com/albums/b263/dewi123sant/capacitor-impedance-curves.jpg[/IMG][/URL]


Why choose 4MHz? Because that is the gain bandwidth of the circuits the PSU1 is designed to power: once the gain of an amp stage has dropped below unity the negative feedback of an amplifier cannot cause regenerative radio frequency pick up. Not that any of our circuits can do this anyway - they are designed to be stable having generous phase and gain margins.

So, the worst case source impedance of the PSU1 output capacitor is 1 to 2 Ohms. Is that sufficiently "short circuit" to prevent radio frequencies to propagate either side of it? Low enough for radio frequencies not to be induced into its 1.5 metre un-shielded DC output cable?

Theoretically it could happen - all cable is inductive, so the further you go from the output capacitor the greater the impedance will be. However, even at the 1.5 metre extreme the impedance will not have risen sufficiently above 1 to 2 Ohms. The RF power in the vicinity would have to be very high indeed (and a danger to health if it were present to that extent) to induce itself onto the wires. You can work this out for yourself from the equation Q (charge) = CV = IT (where C is capacitance, V is voltage, I is current and T is time, the inverse of frequency). So what about lower frequency RF emissions?

As frequency decreases toward the 800kHz minima the impedance drops to around 0.012 Ohms, so we see the situation is much improved.

As the frequency decreases below the 800kHz minima (toward LW radio reception) the impedance rises again but the PSU1 voltage regulator output impedance starts to swamp the capacitor impedance dragging output impedance lower. This can be seen in the following chart from the curve for a similar application which is falling with frequency. So now there are two parallel impedances keeping interference from propagating in the DC output cable.

[URL=http://s21.photobucket.com/user/dewi123sant/media/regulator-output-impedance-curve.jpg.html][IMG]http://i21.photobucket.com/albums/b263/dewi123sant/regulator-output-impedance-curve.jpg[/IMG][/URL]

Also, as the input/output terminals of the voltage regulator are biased apart by the input/output voltage differential, and that voltage is fixed, then the PSU1 reservoir capacitors come into play. These are a quantity of four one thousand micro farad low impedance electrolytic capacitors of the variety used in switching regulators, so they exhibit low impedance at high frequency, and they are arranged in parallel which decreases their impedance four-fold. They will start to swamp the output impedance below 100kHz and will be extremely low impedance all the way down to 40Hz (below hum and ripple frequencies) where the impedance will again rise to about 1 Ohm.

Therefore the PSU1 output impedance is never higher than 2 Ohms and mostly below 0.01 Ohm for its intended use. As such it is a virtual short circuit with respect to the current being drawn from it and signals cannot propagate from a short circuit. Therefore shielding is not required.

Furthermore, the products being powered by it have their own RF power input filters (working at all radio reception frequencies) and are regulated themselves.

Is the PSU1 DC output cable "tough" enough to deliver power to its intended load? Lets take our flagship Solo Ultra-Linear headphone amplifier as the PSU1 load and compare the PSU1 DC output cable to some really serious loudspeaker cable...

To make both examples equal, let the sound pressure level be the same: using a typical 88dB reference sensitivity for loudspeakers and headphones to produce a rather loud 104 decibels!

Heavy duty "audiophile" speaker cable: 326 filaments of 0.1mm diameter, rated at 30 amps, is used to carry a current of 2 amps to deliver 32 watts into an 8 Ohm loudspeaker (to do 104 decibels). That's 15x "overkill".

The PSU1 DC output cable: 13 filaments of 0.2mm diameter, rated at 2.5 amps, is used to carry a current of 0.132 amps when delivering power to a Solo Ultra-Linear headphone amplifier driving both channels at full output into low impedance 32 Ohm headphones (to do 104 decibels). That's 19x "overkill".

The PSU1 DC output cable is therefore relatively more hefty than heavy duty "audiophile" speaker cable!



Graham Slee PSU1 Power Supply'
dewisant
 
Posts: 3
Joined: Thu Jan 02, 2014 10:37 am

Postby dewisant » Mon May 13, 2013 12:00 am

doesnt appear if the pics I uploaded to photobucket work
Attachments
capacitor-impedance-curves.jpg
regulator-output-impedance-curve.jpg
dewisant
 
Posts: 3
Joined: Thu Jan 02, 2014 10:37 am


Return to Audio & Video

Who is online

Users browsing this forum: No registered users and 1 guest