Internet Radio

On Rick’s site he had done a Review on a Internet Radio.

He talks about the Quality of loudspeaker’s – do the sound experts here – have any ideas on better quality replacements speakers for older computers/radios?

How long is a piece of string? ;-)

Without firing up the hifi, I have had a pair of Creative Inspire T10 for many years. These are great for desktop use and I was surprised to see that they are still available! If I had a deeper desk, I might have bought a soundbase like the Cambridge Audio AV2 to double up as a monitor stand.

My radios speech seems more difficult to understand – could be – I am running it at a higher volume and it’s speaker being pushed into a more distorted output.

Generally (within reason), bigger is best.

But, and there’s a very big but, it depends upon the speaker design (you’ll notice speaker cabinets the size of microwave ovens or small fridge actually contain several different speakers), the amplifier, and what you’re trying to play.

The local church has some speakers on the wall that look to be about six or seven inch. It sounds pretty lousy, but for the sort the stuff they play it works.

My radio’s speaker is about three inch. Not great, but better than my desktop powered speakers that are only ~2 inch. I prefer to listen to the internal mono speaker than the stereo powered pair, because there’s just no bass to speak of (even with the bass boost button pressed) in the smaller speakers.

What you lose with size is the ability to push air. So smaller speakers tend to sound tinny, lacking bass and presence. Also, their size means that you can’t get so much sound out of them before it starts sounding bad.

On the other hand, the speakers in the living room are… are they 8" or 12"? I forget. Big and powerful. But while the big speaker can chuck out the bass, it’s not possible to do any justice to the higher frequencies if you’re wobbling like a car crossing a cattle grid. Which is why there’s a passive splitter inside that sends bass to the big speaker, and higher frequencies to a smaller speaker.

I’ll hand over to an audiophile, suffice to say it’s a complicated subject, and what I find acceptable somebody else might find horrible. It’s also partly why the “gold plated plug” trope exists.

Big isn’t always better. I have a pair of IMF Domestic monitors. Each 4 foot tall, 18 inches wide and deep and weighing about a hundred weight.
My wife says there is no room for them in the living room so they have been languishing in the attic for a long time.

I did say “within reason”. ;)

A difficult subject, and yes distortion at high volume is a relatively common issue.

On a personal level I find that the trend to emphasize the high frequencies is a little bit annoying.

Another issue which I have had with my 2 last sets of speakers is that the balance between high and low frequencies changes with the volume setting of the speakers. On the Hercules, it wasn’t too bad I just had to leave it at half volume for a correct balance. On my Edifiers, which are more powerful, I had to set them near full volume and its only thanks to adjusting the output volume of the monitor down to 15% that I don’t have to bring down the volume in DigitalCD to just a few percents.

On a personal level I find that the trend to emphasize the high frequencies is a little bit annoying.

A friend of a friend is a sound engineer, and has worked for quite a few well known bands. He explained that a large number of his compatriots have suffered hearing loss which tends to hit high frequencies first, thus the people doing the mix for live events, and recordings too, emphasise what they can’t hear properly.

This becomes really irritating when it’s a remix of a decades old recording, although frequently it’s more down to the personal prejudice of the engineer mixing. One character comes to mind, where the only time certain instruments are anything other than a murmur in the background is on a recording that they happen to have made, and they were playing that instrument

I’ll hand over to an audiophile

Well, hello there.

(Disclaimer – I’m a ‘sensible’ audiophile, otherwise known as a ‘hi-fi enthusiast’. The main diffference is whereas a hi-fi enthusiast uses their system to listen to music, an ‘audiophile’ uses music to listen to their system.)

So, for my main system, I use a set of Monitor Audio Silver RX6 floorstanders, with matching RX Centre and two sets of RX Surround. I didn’t go with the matching RX-W12 sub, instead I stayed with my old sub, a B&W ASW-1000. I also haven’t bothered going the Dolby Atmos route, partly as 90% of the time I use the system for music, and partly because I don’t want to have to upgrade the otherwise excellent Arcam AVR360 receiver and replace it with one that has Atmos, add another 4-channel power amp for the overheads, and fork out another £400 for a set of four matching Monitor Audio Atmos in-ceiling speakers.

The bedroom system is a bit more modest, a Yamaha RX-V673 running 5.1 channel bi-amped ¹. Speakers are Gale 5 and Centre 2 fronts, and the rears have recently been upgraded to a set of Gale 3060B bipolar dedicated surrounds (£25 off eBay), along with a Paradigm PS-1000 sub.

Where was I? Ah, yes…

Generally (within reason), bigger is best.

Depends on the speaker material. The more rigid the cone, the bigger the surface area can be made without sacrificing what hi-fi magazines like to call “agility”.

For bass/mid speakers (not subwoofers etc) it’s generally better to double up the drivers rather than fitting a bigger drive unit. My Gales do this, they have a pair of 5 inch bass-mid drivers running in parallel (12ohm drivers so the speaker presents a 6ohm impedence to the amplifier).

¹ I mentioned bi-amping. The Yamaha receiver in the bedroom has 7 channels of amplification. There’s an option in the setup menus to configure this for “EX surround”, “Zone 2” or “Bi-amp LR”. It’s set to the third position, where channels 6 and 7 power the tweeters, channels 1 and 2 power the bass-mids. On the Yamaha it does make a noticable difference when playing music; more detail, better soundstaging, in other words all the ‘twinkley’ bits are more audible. Tried it on the Arcam and it made no difference at all, I can only assume the Arcam’s output stage can deliver a lot more current so is less affected by speaker loading.

The RX6 speakers in the front room have an interesting arrangement, they’re configured as a 2.5 way design. There’s 3 drivers, tweeter, bass-mid and bass. The tweeter and bass-mid are basically identical to the RX1 model, but the lower part of the cabinet has the bass driver mounted in it. This has an internal baffle so is effectively a separate enclosure. Interestingly the lower bass enclosure is ported to the front of the cabinet, the upper bass-mid is ported to the rear. Monitor Audio supply some foam plugs with the speakers that can be stuck into the ports to reduce the bass – I’ve found with my setup it sounds best if the rear (top) ports have the foam plugs in them, with the lower front ports being left open.

The Bronze range had a similar arrangement with front and rear ports.

The most scientific test I can come up with is to play a pink noise tone through the system and use the (free) spectrum analyser app on my phone. As expected it comes up flat from about 25Hz right up to just over 20kHz.

It’s also partly why the “gold plated plug” trope exists.

Actually there’s a very good reason for gold plated plugs. Gold doesn’t tarnish or oxidise.

So imagine you have a plug where the pins aren’t gold plated (nor are the socket pins). There’s a thin film of oxide builds up on the contact surfaces. When you mate the plugs together you end up with two contacts (conductors) separated by a thin film of non-conducting oxide (dialectric)…

Congratulations. You just made yourself a capacitor.

This doesn’t happen with gold-plated contacts. No point using 24ct gold though, it’s too soft. A simple electroplating of 9ct gold will stop any oxide forming.

No need for expensive plugs either. Farnell / CPC sell 4mm gold banana plugs for about 79p a pair (red and black).

I read “gold plated plug trope” as “directional cable” or “oxygen free copper”. Gold plating is a good thing. Less crackly.

Ah yes, the myth of directional cables. A myth which started because some professional cables are marked with an arrow to show the direction from the source to destination, which is a very useful thing in a studio (or at a live event). Makes no difference to the sound of course, it’s just for identification.

Of course the ‘audiophools’ latched onto it, and now ‘cables must be directional’.

One of the things that makes me laugh is “gold plated TOSlink plugs” to “improve contact reliability”. Not sure if I’m the only one who can see the problem there?

Also… ‘audiophile’ USB and Ethernet cables? Ummm…

Well, gold is very shiny. Stands to reason. ;-)

Iridium-plated terminators: if it was good enough for the dinosaurs…

BTW if you haven’t googled ‘Chicxulub crater’ recently, please do so.

BTW if you haven’t googled ‘Chicxulub crater’ recently, please do so.

Oh, very good.

Ah, so you actually have to use Google! DDG didn’t show anything unexpected, so I was wondering which article was in question.

Yup. I’m normally a DDG user, too.

There’s similar easter eggs if you google the Friends characters.

This doesn’t happen with gold-plated contacts.

Sorta – I believe most units don’t have gold-plated sockets, so the socket side can provide you a good line in oxide layer.
An oxide layer that will abrade away on removal/insertion provided the item being inserted isn’t nice soft gold.

Not sure if I’m the only one who can see the problem there?

My first thought is that the gold needs to be very, very thin otherwise it’s going to be a bit of an optical problem :)

The one that really gets me is the silly cost on the audio fibre patch leads compared with the data fibre patch leads that have a more demanding bandwidth requirement.

Also… ‘audiophile’ USB and Ethernet cables? Ummm…

That reminds me of the occasion a supplier was suggesting that the images appearing on the PACS viewing screen upside down was “a network problem”
I replied that they would need to figure out a method of dealing with the effect of the twist length of the pairs in the patch cables, as we couldn’t afford the time to trim their length to keep the number of twists to a full 360 twist at the receiving device.
It took them a couple of weeks to figure out that I was, er, “extracting the urine”

Back to (sort of) the original topic…

On my main desktop PC I’m using a set of Altec Lansing 121 PC speakers (two desktop units with a separate bass unit under the desk – marked as a ‘subwoofer’ but trust me, there’s nothing ‘sub’ about it) for the “system” sound, I also have a PCM2704-based USB sound card linked up as a secondary card. The optical output of this runs to a Sony micro system (CD / cassette / MD / tuner, picked up for £12 from a charity shop as it has an optical input that can be used as a digital ‘line in’) driving a set of Celestion 1 speakers (an old set I had from my student days, small but sound decent – far better than the ones that came with the Sony unit).

The speakers have to be plonked straight on the desk, so I’ve attached some rubber feet to the bottom.

I’ve then set up the default Windows audio device as the on-board sound (which feeds the Altec Lansings) and Winamp, Foobar2000, Audition etc use the USB audio. That way I can crank up the music without being deafened when Windows pings for a new email etc.

This doesn’t happen with gold-plated contacts

Be careful about gold plated contacts. Gold on both contact surfaces is of course good, but gold against tin will fail within a few years if not moved.

I originally came up against this problem in a joystick for a Sinclair Spectrum. The box proudly proclaimed fully gold plated contacts. One direction of the switch eventually failed. The culprit was one of the crimp connectors, which was indeed fully (not selectively) gold plated, and had ceased to make contact with the tinned copper wire to which it was still perfectly well crimped. (The solution was to flow solder over the crimped connection.)

I came across the same problem a while later with Molex KK connectors. We had been rather lax in specifying selectively gold plated female contact springs, but male pin headers, and some of the pairs lost contact. All that was necessary was to unmate and mate them, whereupon they would work again for a while.

Tin against tin is more reliable than gold against tin.

The problem is known, although not widely so.

So, if you’re using gold plated contacts, make sure that both male and female contacts are gold plated – otherwise you’re actually better off with neither.

So, if you’re using gold plated contacts, make sure that both male and female contacts are gold plated – otherwise you’re actually better off with neither.

Indeed. Not had to worry about that at work for a couple of decades though. S.E.P these days.
Then I had to be at the kit to make things work again, these days I’m normally not within miles of the stuff I’m working on.

Interesting point about gold-tin contacts. Unsurprising now it’s pointed out, but not something I’d ever come across or thought about. Thanks folks!

Related to something I’m very conscious of in other contexts… http://clive.semmens.org.uk/Fiction/Penny/Police.html#AluRot

The one that really gets me is the silly cost on the audio fibre patch leads compared with the data fibre patch leads that have a more demanding bandwidth requirement.

I’ve always ordered Toslink cables from Farnell / CPC. They cost about a quid each for a 1m cable. The bandwidth on SPDIF (whether Toslink or 75ohm co-ax) is about 1.5Mbit/sec. I really can’t see how “audiophile” grade optical interconnects could make a damned bit of difference.

Also I once bodged together a co-ax to optical converter by attaching some wire to a 5mm red LED and using blu-tak to fix it into the optical input on an AV receiver. Worked great as a proof of concept. I later did something slightly more elegant involving boring out the centre of a Toslink plug and fitting a 3mm flat-top red LED into the housing with a bit of screened cable going to a phono plug on the other end. Looked professional (ie not like a bodge) and worked perfectly. Still working now.

Be careful about gold plated contacts. Gold on both contact surfaces is of course good, but gold against tin will fail within a few years if not moved.

I think most non-gold audio connectors use nickel, which is pretty robust. Not as tarnish resistant as gold though.

I originally came up against this problem in a joystick for a Sinclair Spectrum. The box proudly proclaimed fully gold plated contacts. One direction of the switch eventually failed. The culprit was one of the crimp connectors, which was indeed fully (not selectively) gold plated, and had ceased to make contact with the tinned copper wire to which it was still perfectly well crimped. (The solution was to flow solder over the crimped connection.)

Which is why you always solder speaker or line plugs onto the cable, and don’t crimp them.

I have to say I’ve never had a problem with gold-plated plugs and nickel-silver sockets (or vice versa). The gold plating is 9ct – anything more than this would be pointless and wouldn’t have the hardness to stand up to repeated insertions (stop laughing at the back there!).

Of course some ‘audiophools’ would have it that 24ct gold is ‘superior’. Most of whom don’t realise that silver is actually a better conductor, the only reason it’s not used is because silver does tarnish.

The ultimate speaker cable would of course be silver-plated oxygen-free copper with gold-plated plugs.

I’ve always ordered Toslink cables from Farnell / CPC. They cost about a quid each for a 1m cable. The bandwidth on SPDIF (whether Toslink or 75ohm co-ax) is about 1.5Mbit/sec. I really can’t see how “audiophile” grade optical interconnects could make a damned bit of difference

Exactly, mega amounts of bovine effluvia coating that kit, never mind any gold.
Even better is the damn stupid 0 degree symmetry of the connector – it’s f’n round fibre that just might mate the face of the socket optic with the face of patch fibre optic a little better when rotated 90 degrees (or 180, or a miniscule 2 degrees.

The ultimate speaker cable would of course be silver-plated oxygen-free copper with gold-plated plugs.

Producing a difference in sound only appreciated by the local cats, dogs and bat population.
There’s more negative effect in having someone like Steve Wilson re-mix a Yes album. (Select your own preferred music and inept¹ mix “engineer” as appropriate)

¹ Not that I believe SW’s bad mixing is a lack of talent in that department, more an expression of jealousy and mixing down better musicians content)