New Posts New Posts RSS Feed - The Audio Science Review Forum
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Welcome to the Graham Slee Audio Products Owners Forum

 

Open to all owners plus those contemplating the purchase of a Graham Slee HiFi System Components audio product and wishing to use this forum's loaner program: join here (Rules on posting can be found here)

This website along with trade marks Graham Slee and HiFi System Components are owned by Cadman Enterprises Ltd


The Audio Science Review Forum

 Post Reply Post Reply Page  <1 910111213 19>
Author
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 12 Jan 2021 at 8:40pm
Hello all,

David and I have been having a private discussion. Still, perhaps it would be better to throw it open to all interested members.

I have resisted joining in on this topic because I think it a waste of energy engaging with those easily-led by those of poor understanding.

I wish vinyl was as easy to understand as the uneducated believe it to be. Sadly, it is considerably more complicated, so much so that the examiners of the US and UK patent offices had difficulty because they had to learn and re-learn RIAA EQ to comprehend my patent. The US examiners got there early, and by the looks of things, the UK examiners are going to take their word for it rather than go through the whole rigmarole.

I have tried to explain that the output of a generator rises with increasing speed. The cartridge is the generator, and the frequency is the speed. Without a regulator, the output of a generator increases. It should be obvious to the older end here, who ever owned an old car, found the bulbs blackening and blowing, found the battery had bubbled over, and then realised they needed a new regulator on the alternator.

It should also be evident to those old enough that the ceramic cartridges of the good old days gave an almost flat response. Still, when a magnetic cartridge was connected to the same input, save for the lack of volume, the sound contained minimal bass, not much mids, but tons of highs.

The type of input I mean is the 1 meg or larger grid input resistor of a valve (tube) input. An input that has no EQ - just plain and simply flat. Yes, some sounded dull because there is record EQ. Still, only a dip between 500Hz and 2122Hz - otherwise, the record is recorded flat: it's called constant amplitude.

When played using a generator - a constant velocity device - a magnetic cartridge, whether MM, MC, or MI, the flat becomes an upward curve. The dip becomes flat. The flat upper portion again becomes an upward curve.

The curve is not homogenised. The curve is two separate curves happening at different times in history. 1. when the record was cut; 2. when you play the record.

I shall not delve into the 50Hz lower frequency pole, or zero, depending on which way you look at it, right now. The point is that the flat gain stages beloved by hi-fi marketers have considerable headroom at low frequencies and virtually none before their passive interstage filters cut. You can have one doing the bass and another doing the HF. The HF one gives you the chance of claiming you're god, at 1kHz.

So then, what about the clicks and pops? Believe it or not, they're fast spikes, and therefore they're high frequency. If they don't cause the record to skip, they're below cutover amplitude, and that's approximately 5 times the recorded level. If you don't have any headroom left up there, the result is a gain stage out of control because its NFB has no source anymore. And to those who say there's no NFB, there always is. It's just the usage of words to explain a different type.

An active stage, where the EQ network is the negative feedback, provided there's sufficient loop-gain, has a constant overload margin. Just how much is required depends on understanding the recording process and the cutover ratio. A cartridge doesn't reproduce music (as such) when skipping - when it's stuck, and that's cutover. There are known limits, and most vinyl can only reach +14dB (5 x the signal) on music. You can cut grooves at higher amplitude but only at specific frequencies and waveforms - that's what they record on test records.

If the limit is 5 times, you don't need any more headroom (or, as some call it, saturation).

As I wrote to David earlier: "I choose to use active EQ because it results in a constant overload margin. In passive EQ, the overload margin narrows with frequency, but if you decide to advertise "the right end," you can hugely impress your customers."

However, active EQ comes with difficulty, and it can result in clicks and pops just as loud as a passive interstage equaliser. The reason is overshoot, which is to do with the propagation timing of the stage. It has plenty to do with slewing-induced distortion. The subject was initially discussed (regarding power amplifiers) in an AES paper by Matti Otala in 1972.

If you can imagine the input as a "singleton" transistor, or even a differential pair, if it has no emitter degeneration, it overloads and passes the signal directly to the "voltage amplifier" section. However, that is Miller stabilised, which means it must charge the Miller stability capacitor. Modeled passively, it can be seen that the capacitance isn't the value of the capacitor, it is magnified by the gain, and so it represents a very awkward load to the source (the cartridge). Most transistor stages were like that because by adding emitter degeneration, the hiss increased considerably.

Now, also, you have the EQ in the negative feedback trying to cancel the noise, but you can see that due to the Miller capacitor, there's a delay. The NFB sees a fake error voltage and corrects what should not be there. This takes several cycles to dissipate, during which the stage clips its power rail and extends the click, making it a thick click.

The above is only scratching the surface (pardon the pun), but I'd like all to know that my own learning started in public entertainment (playing records to audiences) more than 45 years ago. Now approaching my state pension (nearly 66), I do get a bit pee'd off about know-all upstarts and don't really want them publicised in my space.

I might talk a little about testing next time I'm on.
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 13 Jan 2021 at 7:09am
They all eventually turn. Some start that way. The Solo sounds really bad, said Mr. G on Head-Fi, later admitting he never had one nor heard one, his excuse being he was an X-Can enthusiast. That's right, Mr. G, kick the working guy in the nuts. He also has a forum.

It must be really tough for the Chinese. After drubbing mine, almost always and without fail, the Cambridge phono saves the day! Audio Scaremonger Rer, you are not the first, and you won't be the last. Same with Google!

Not that I want to turn this into a rant about the Chinese Communist Party, but many fingerprints make light work. It seems Big-Tech controls "free speech" these days.

You'd think they'd leave the harmless working guy alone. Seems they need another 3,000 sales to keep afloat (poor buggers), either that or they have blind-mans sn*t syndrome (must I explain?).

I'm such an amateur, you know! "he has no factory - look, this is his registered address - he works from home, haha."

Sure, I don't have a million square feet with dorm on top or suicide nets, and I try very hard not to support Uiger Muslim torture. I'm a guy who learns from his past mistakes.

You can view the industrial unit on a South Yorkshire industrial estate - it's still there - but with a different occupier.

Alan, the manager, was taken aback when I told him what the three pipes emerging from the wall had been used for (compressed air). He was somewhat gobsmacked as I showed him around his premises, giving him its history, but that history was from the 1980s.

There is an "about" page, you know? I guess nobody ever bothers to read it. It might seem to some that it's just made up on the fly; after all, there are no references you can check. Well, the one I've at last been able to mention - Audionics Sheffield Ltd - went bust a few years ago, complete with "director shame," so can no longer leg me down with legal threats. They were once the engineering arm of YRN PLC, who owned Radio Hallam, Pennine, and Viking (plus Magic AM), who was bought out by Metro, cutting my employment short! I was a senior engineer.

The other two are a bit dodgy; even the mention of Station Lane could see me in court. However, I might get away with Chevin Research because Martin always had a headache (he will know what I mean). Then there are the little jobs in-between I could boast of, but on second thoughts, that's just petty.

Beyond the cover of Google street view (TM), which I'm told ASR shows, lay my 300 sq ft workshop, and in other parts of South Yorkshire and the North Midlands, the additional 500 sq ft exists. Provided it's legal, you can form a loose workers cooperative in the UK - how very working-class and non-communist Chinese can we get?

We are, in fact, three separate companies. John and Leo run theirs, and I run mine. I am 100% HiFi System Components Ltd, and its sole director. They have their own "clientele," amongst which is HiFi System Components Ltd. We used to be Graham Slee Projects Ltd until "World politics" took out 70% of its customer base. We agreed to part on good terms in 2015 and came back together as a workers cooperative. HMRC punished me for phoenexing. I paid the bill, and since then, they've been giving it back.

Go check ASR!

https://companycheck.co.uk/company/09719303/HIFI-SYSTEM-COMPONENTS-LIMITED/companies-house-data

https://companycheck.co.uk/company/05637059/GRAHAM-SLEE-PROJECTS-LIMITED/companies-house-data

https://companycheck.co.uk/company/10417687/GRAHAM-SLEE-PROJECTS-LIMITED/companies-house-data

In the UK, we're totally transparent about these things, unlike in the US, China, and many other countries, where they can easily pull the wool over your eyes. Now, where is ASR based?
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Sylvain View Drop Down
Senior Member
Senior Member


Joined: 18 Jan 2010
Status: Offline
Points: 481
Post Options Post Options   Thanks (1) Thanks(1)   Quote Sylvain Quote  Post ReplyReply Direct Link To This Post Posted: 13 Jan 2021 at 2:07pm
Thank YOU very much for the clarification ....and the product harmonise well with my sense of hearing i  terms of sonic timbre and delighted to be A CUSTOMER to a 'Workers Co operative''' and very grateful of the research base product that provide good insights rather the Salesman hipe to part my funds. To derive good Synergy from the overall system, read the forum information again, and an email to GSP always guide and for free. Tweaking no longer ump hazardous!!!!
Back to Top
patientot View Drop Down
Senior Member
Senior Member


Joined: 28 Nov 2018
Location: USA
Status: Offline
Points: 1523
Post Options Post Options   Thanks (0) Thanks(0)   Quote patientot Quote  Post ReplyReply Direct Link To This Post Posted: 13 Jan 2021 at 2:12pm
Originally posted by Graham Slee Graham Slee wrote:

Now, where is ASR based?

The person that posted the Google Street View comment is a new member with only one post to their name over there. 

I guess they didn't see the irony that ASR also measures equipment out of their house and that the proprietor (a retired software engineer IIRC) also runs a home theater installation company out of his house, with another space in a shopping center. Based in Bellevue, WA, USA. 





SL-1200 MK7 (modified) + Reflex M + PSU-1 used with AT150-40ML, AT VM95ML, Stanton 680mkII + Ogura, and Shure M35X cartridges.
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 13 Jan 2021 at 2:47pm
As promised, some information on measurements.

Every audio analyser has a signal generator (oscillator) which outputs the frequencies and amplitudes required for the device under test (DUT), this example being the Reflex C.

All signal generators have output impedance, or should I say output driving impedance. On AP's ATS-1 machines, the lowest they could go was 50 ohms. On the APX525, AP managed to reduce output driving impedance to 20 ohms.

We use both of these machines. We have to understand the attenuation a low impedance input causes to the analysed measurements.

If the ATS-1 is used, a stereo load of 100 ohms (the standard input on a Reflex C) is seen as 50 ohms by the oscillator output - which drives both channels in parallel. As its output driving impedance is also 50 ohms, then the Reflex C output is half what it should be (-6dB). That's because its input is half what the ATS-1 tells us. The Reflex C gain, however, is still correct.

On the APX525 with 20 ohms output driving impedance, the Reflex C output is 71.428% of what it should be because the input it is receiving is 71.428% of what the APX525 says. Still, the Reflex C gain is correct.

The only way to test a Reflex C properly is to lift one end of each input load resistor such that the test set's oscillator isn't overloaded. Experienced testers know this stuff, and when sometimes they forget, you usually hear them exclaim "finger trouble" or worse, as they realise what they've done.

An oscillator driving a heavier load than it's designed to do will distort. That distortion adds to the distortion reading, and so the inexperienced tester will record a false reading.

A distorting oscillator might also become noisier and drift off frequency, rendering a false FFT trace.

A word of warning for anybody using the loaner program to facilitate such measurements. If the product gets disassembled in any way, we will know about it. Independent test labs will either agree to pay a nominal fee or even purchase the product.
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
patientot View Drop Down
Senior Member
Senior Member


Joined: 28 Nov 2018
Location: USA
Status: Offline
Points: 1523
Post Options Post Options   Thanks (0) Thanks(0)   Quote patientot Quote  Post ReplyReply Direct Link To This Post Posted: 13 Jan 2021 at 3:25pm
Graham, I'm no engineer, but I think this must explain the problem. Looks like a 20 ohm output was used which recorded the gain at 59.2dB instead of 62dB. I actually thought they might have been testing an older version of the Reflex C because the gain didn't match current specs. Looks like it was just tested wrong.

Here's my layman's arithmetic. 

59.2db = a gain factor of 912

The correct gain factor should be 1260

912 divided by 1260 = 72.38%
SL-1200 MK7 (modified) + Reflex M + PSU-1 used with AT150-40ML, AT VM95ML, Stanton 680mkII + Ogura, and Shure M35X cartridges.
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 13 Jan 2021 at 4:09pm
Or roughly 3dB down which is 0.707 - close enough.

In the UK, AP can boast James Kelly

Quote AP: "James Kelly is a British test and measurement engineer with over 20 years' experience of designing custom automated test solutions for a variety of manufacturers in the audio industry, mainly using AP audio analysers. Before joining Audio Precision, James worked for companies including ARCAM (where he helped to develop the first ever functional 'bed of nails' testing facilities using AP analysers), TAG McLaren Audio, and International Audio Group (IAG) where he served as Head of Engineering in both Cambridgeshire and Shenzhen, China. Prior to leaving IAG, he designed automated test facilities for the manufacturing groups of Quad, Wharfedale, Mission and AudioLab."

James is quite happy in having an afternoon's testing session with me when he's in the area. Last time was before the China Flu. I think he enjoys working with knowledgeable customers. I have a lot of respect for James. It would be nice to give him more custom, and upgrade the ATS-1 with another APX525. Unfortunately, with such ill-informed "press" we don't get the turnover to justify it. Just an opinion, but it often looks like those sites are very biased against us, and have some axe to grind against honest working people.
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
 Post Reply Post Reply Page  <1 910111213 19>
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.01
Copyright ©2001-2018 Web Wiz Ltd.

This page was generated in 0.156 seconds.