REVIEW - Pheonix Engineering PSU + Tachometer

[Wakefield Turntables]

I take it you are feeding the supply from the P10, not the other way around :-)

Yep!

[struth]

I found a regen made a fair bit of improvement with my thorens. Pretty night n day

[Wakefield Turntables]

I wonder if the update rate can be improved by adding additional magnets on the underside of the platter, I've just emailed Bill to find out. I wonder if this would improve pitch stability.

[lurcher]

I wonder if the update rate can be improved by adding additional magnets on the underside of the platter, I've just emailed Bill to find out. I wonder if this would improve pitch stability.

You would have to get the magnets exactly spaced around the platter, not a issue with just one and setting the average speed to 33.3, but next to impossible to do with more than one. You would need a rotary encoder or a toothed wheel frequency generator like direct drives use. You would then face the issue that the motor and platter is designed to work at a fixed speed driven by the AC motor, you could not drive the phase of the 50Hz and expect to see control like a direct drive system.

Not that I have ever heard a problem with pitch stability will a well fettled garrard. I would expect a good 50Hz supply would work wonders which is what the supply you have should provide. I am messing with such a thing myself, you can compare it to the one you have if Its done by the time the preamp is ready.

You can see how the UK mains supply frequency varies here: http://www2.nationalgrid.com/uk/Indu...erational-data, and I am sure a lower distortion sine will help reduce motor vibration. It was interesting looking at the mains on a scope compared to a quartz referenced 50Hz and see how it wanders around the target value.

[Wakefield Turntables]

You would have to get the magnets exactly spaced around the platter, not a issue with just one and setting the average speed to 33.3, but next to impossible to do with more than one. You would need a rotary encoder or a toothed wheel frequency generator like direct drives use. You would then face the issue that the motor and platter is designed to work at a fixed speed driven by the AC motor, you could not drive the phase of the 50Hz and expect to see control like a direct drive system.

Not that I have ever heard a problem with pitch stability will a well fettled garrard. I would expect a good 50Hz supply would work wonders which is what the supply you have should provide. I am messing with such a thing myself, you can compare it to the one you have if Its done by the time the preamp is ready.

You can see how the UK mains supply frequency varies here: http://www2.nationalgrid.com/uk/Indu...erational-data, and I am sure a lower distortion sine will help reduce motor vibration. It was interesting looking at the mains on a scope compared to a quartz referenced 50Hz and see how it wanders around the target value.

Bugger that then!

[paulf-2007]

I was interested in these some time ago when I had several idler drives and didn't get around to buying one. It certainly looks a nice bit of kit but I had to smile when you said it stabilises in a minute. The Denon DP80 gets up to speed in 0.9 seconds. Admittedly the speed stability is not as accurate at 0.002% but then it's technology is 40 years old.

[daytona600]

I was interested in these some time ago when I had several idler drives and didn't get around to buying one. It certainly looks a nice bit of kit but I had to smile when you said it stabilises in a minute. The Denon DP80 gets up to speed in 0.9 seconds. Admittedly the speed stability is not as accurate at 0.002% but then it's technology is 40 years old.

PSU limited by the drive train in idler & belt drive turntables

comparism between PLL ( Phase locked loop ) & DDS ( Direct Digital Synthesis )

add the matching tacho this forms a closed loop speed control system for 33.000 & 45.000 rpm



Direct Digital Synthesis is a digital frequency generator that uses a crystal controlled master clock to increase the count in an accumulator register by a fixed amount on each clock transistion. The register "accumulates" the count on each clock which increases linearly, until the maximum count is exceeded, the counter overflows to zero and the processes begins again and continues indefinitely. The amount the accumulator changes on each clock input is determined by the frequency control register: The larger the number, the faster the count accumulates, the lower the number, the slower the output changes. The accumulator contains a binary number between 0 and 2^N where N= the width of the register in bits. DDS accumulators are usually 24, 28 or 32 bits wide. The smallest number it can increase by is 1, so the frequency output resolution becomes MClk/(2^N); the higher N is (or the lower the Master Clock frequency) the finer the resolution. The output frequency is Freq Count x (MClk/2^N). For example, if the crystal frequency (MClk) is 4.096 MHz and the accumulator is 24 bits, the frequency resolution is .244140625 Hz (~ a quarter of a Hz). For 60Hz the Freq Count=246 (60.0585 Hz). If the crystal reference is divided down to 32kHz (÷128), the frequency resolution is 0.001907 Hz and Freq Count for 60Hz becomes 31,457 (59.9995 Hz). The same MClk with a 32 bit accumulator has a frequency resolution of 0.00000745 Hz and Freq Count for 60Hz becomes 8,053,064 (60.0000024 Hz).

The highest order bits of the accumulator are used as an address in a look up table that converts the linear phase accumulation into a sinewave amplitude value. This value is then sent to a digital to analog convertor (DAC) which produces an analog sinewave output. The DAC resolution is typically between 8 and 14 bits.

Because DDS is a feed forward system, it doesn't require a feedback loop and frequency changes take place on the next MClk input, are phase coherent and settle to the new frequency in uSecs. The output frequency will have the same stability as the crystal reference clock and phase noise will be better than the crystal by a factor of MClk/N where N is the frequency count. DDS is a type of sampling convertor, so the maximum frequency output is MClk/2 (Nyquist limit), but in practice, the output is usable to ~MClk/10. The closer the output frequency is to this limit, the worse the noise and spurious performance will be. Analog filters at the output of the DAC can reduce the noise and spurious content to low levels.

DDS is usually done by dedicated ICs that are complex to program and configure. A microprocessor is needed to compute the frequency count and perform the programming and communications with the DDS chip. DDS is commonly used in bench top test equipment and high end signal generators.

[Wakefield Turntables]

I was interested in these some time ago when I had several idler drives and didn't get around to buying one. It certainly looks a nice bit of kit but I had to smile when you said it stabilises in a minute. The Denon DP80 gets up to speed in 0.9 seconds. Admittedly the speed stability is not as accurate at 0.002% but then it's technology is 40 years old.

It stabilises usually within 10-15 seconds due to the fact that the tachometer updates the PSU every 1.8 seconds and not every few milliseconds like all DD's. It is a nice piece of kit, you should buy one.

[paulf-2007]

It stabilises usually within 10-15 seconds due to the fact that the tachometer updates the PSU every 1.8 seconds and not every few milliseconds like all DD's. It is a nice piece of kit, you should buy one.

Its not something I could use with the Denon DP80. It has its own built in speed stability
Who knows what the future holds, never say never

[al2002]

Hello Rexton. Nice report!

A couple of questions:

1. Is this a two phase or one phase supply? If two phase, is the phase angle between the two phases adjustable?

2. Does the supply drop the operating voltage relative to the startup voltage?