Power Conditioners

Furman IT-Reference 20i Power Conditioner

ARTICLE INDEX

Furman IT-Reference 20i

Introduction

I’ve been suspicious of power conditioners ever since they first appeared – sometime in the Jurassic Park period I think. Just like the science in that movie is a little glossed over, so is the science of AC power glossed over by wild claims for power conditioning products. Besides, when we shell out the big bucks for good hifi a lot of what we’re paying for is the power supply. That’s why we pick it up before we buy it. A good power supply means a big transformer, which is heavy. And big capacitors which we’d like to see neatly arranged. I was willing to give Furman a try however because for one thing, they got their start in pro-audio.

I don’t necessarily think that the pro-audio world has an advantage over the retail audiophile crowd (that’s you and me) when it comes to good ears. After all, these are the people who destroy most recorded music with godforsaken compression. And, I’ve read more than one interview with sound engineers denouncing audiophiles as foolish wasters of money. True enough if all we have to listen to are those overly compressed recordings. Still, there are also some famous audiophiles amongst the philistine sound engineers and the pro-audio world does pay special attention to power. Another kicker for Furman, I saw a their power conditioners being used by more than one hifi manufacture at last year’s Rocky Mountain Audio Fest.

I was willing to give them a try, but still I was suspicious. What can be done for the power supply that isn’t already being done in my expensive components? My current CD player is a Naim CDS2 which comes in two boxes, the CD player (transport and DAC) and the power supply. When this thing was new in 2002 it cost $11,000. I would hope that at that price there isn’t much more that could be done in either of the boxes. Well there may have been something because Naim followed up with the CDS3 and CD555, the latter of which costs about the same as a nice car.

Specifications

  • Input: 20 Amp Circuit Required
  • Output 12-20 Amp Continuous RMS (All outlets combined)
  • Linear Noise Attenuation:
  • Transverse (Differential) Mode:
  • >20 dB, 1kHz to 2 kHz
  • >40 dB, 2kHz to 100 kHz
  • >80 dB, 100 kHz to 1 GHz
  • Common Mode:
  • >95 dB, 10 Hz to 50 kHZ
  • >40 dB, 50 kHz to 1 MHz
  • Outlets: 8 Symmetrical Balanced
  • 4 Power Factor Technology
  • Transient Voltage Surge Protection
  • Telco – Cable/Satellite: Less than 1dB Line Loss
  • Dimensions: 5.75” H x 17” W x 15.625” D
  • Weight: 83 Pounds
  • Power Consumption: 8.5 Watts
  • MSRP: $3,499 USA
  • Furman Sound

 

In spite of my suspicions there were two things that I was sure a good power conditioner could do without stretching the imagination. One is to filter the AC effectively for high frequency noise. There’s plenty of it and more every day. This is something that is done to one degree or another by all power conditioners, even by budget power strips. Furman takes some steps to make the filtering linear – that is, even across the frequency spectrum. The other tried and true method for power supply conditioning that I was familiar with was to create balanced AC.

The 120 V AC mains are run to a transformer, the transformer is wound to provide a +60 and a -60V output. These balanced signals become the basis for the mains supply to the components. This is something that is surely not common. Even amongst the Furman line of products only the IT-Reference 20i performs this trick. It’s expensive to do, requiring a large, specially built transformer. The IT-Reference 20i weighs 83 lbs so I have no doubt that there’s at least one large transformer in there.

There is one other power supply conditioning trick which is no trick at all – simply regenerate the AC with a dedicated power amp. The inefficiency of this has never appealed to me and it makes it hard to apply to the power amps. Build a power amp to power the power amps? How big would that thing have to be? There’s got to be a better way. Some people are now trying it by making the conditioner from efficient Class D amplifiers. Ok, that’s worth a try. I have yet to hear that particular trick. Furman doesn’t do it.

Those are the power conditioning tricks I knew of, the IT-Reference 20i has a couple more. Furman has something called ‘Power Factor Technology’. That’s a little bit like claiming to have ‘Electricity Technology’ as all power circuits have an associated power factor. It has to do with the impedance of the power supply and the reactance of the load. Similar to your when your power amp has to drive a speaker with an impedance dip at certain frequencies, the source (power amp) is required to produce much more power than is required at other frequencies because of the nature of (reactance of) the load.

In theory, a good power amp is designed to be able to supply whatever current is necessary even for highly demanding speakers that might dip to as low as 2 ohms at some frequencies. Specifically it is the large capacitors that are meant to be the reservoir of power (current) for this situation. Furman’s claim, if I understand correctly, is that even if they are humongous capacitors, able to supply large amounts of current for short bursts, when they are drained quickly the DC supply that your amp’s power supply is trying to provide will dip down ever so slightly and the transformer and AC mains will be called upon to supply an equally large current to maintain the designated rail voltage. This temporary requirement for large amounts of current is the reason that aftermarket power cords might make a difference in some situations. It’s the impedance of the power cord that is making the difference. The Power Factor circuit in Furman conditioners seek to solve this problem with a ‘tuned reactance circuit’ made up of capacitors able to supply peak currents of over 80 amps.

Well, enough theory. I could go on, if I only I knew more. Impedance is fun to study (I’m serious, really) but if you haven’t looked at it in 20 years, memory is impeded. The proof of the audio component is in the listening after all.

Furman IT-Reference 20i