New Posts New Posts RSS Feed - 1970s Design Indulgence
  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


1970s Design Indulgence

 Post Reply Post Reply Page  <1 2122232425 345>
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 (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 15 Oct 2018 at 1:13pm
Originally posted by Richardl60 Richardl60 wrote:

Like most of us I think I sort of follow this thred......
 
From the discussions above re heat/heat sinks does this perhaps indicate why some of the monstrously heavy 'class A designs' from the later 1970s-80s all seem to have huge heat sinks or fans (thinking here of say the older Threshold/Krells) to keep the little darling components suitably cool?


Correct! Without such precautions you'd end up with a blown amplifier or even worse, a fire!


Edited by Graham Slee - 15 Oct 2018 at 1:13pm
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 (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 15 Oct 2018 at 7:14pm
Here is the (hopefully) final "1970s power amp" and its measured spec.

50W Power Amplifier Circuit Schematic

Measured Spec.

Mains voltage: 230VAC

Dummy load: 2 x 8 Ohms exact, resistive.

Test analyser: Audio Precision APX525

Input sensitivity: 763mV = 0dB this test

Results Left/Right (rounded up or down: actual results were slightly better than shown)

Output stage quiescent current (Iq): 16.2 mA/17.4 mA

S/N: 74dB/74dB ref -28.375dB input, 20Hz - 20kHz unweighted

S/N: 74dB/74dB ref 0dB input/28.375 dB gain, 20Hz - 20kHz unweighted

Frequency response: 12Hz - 50kHz -0.5dB ref: 50W and 1W

Intermodulation (IMD, SMPTE): 0.2%/0.2% ref: 50W; 0.065%/0.07% ref: 1W

THD+N (300kHz low-pass filter engaged):-

1W (-17dB input): 0.035%/0.035% at 1kHz; 0.155%/0.150% at 10kHz; 0.27%/0.26% at 20kHz

50W (0dB input): 0.055%/0.055% at 1kHz; 0.16%/0.16% at 10kHz; 0.57%/0.55% at 20kHz

Crosstalk: -62.5dB

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 (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 15 Oct 2018 at 7:29pm
Forgot to add that it was both channels driven, and each test allowed the APX525 to null-out fully, and so it wasn't a quick snapshot, more like several minutes, and so it should have said both channels driven continuously.
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Richardl60 View Drop Down
Senior Member
Senior Member


Joined: 04 Nov 2014
Location: Yorkshire
Status: Offline
Points: 1468
Post Options Post Options   Thanks (0) Thanks(0)   Quote Richardl60 Quote  Post ReplyReply Direct Link To This Post Posted: 15 Oct 2018 at 10:26pm
Does output vary much with load Graham?
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 (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 16 Oct 2018 at 3:40am
Originally posted by Richardl60 Richardl60 wrote:

Does output vary much with load Graham?


I haven't got a clue because every time I ran it into 4 ohms the output fuse blew at just above 70 watts. I only had 2A fuses and 4 ohms indicates I should use a 5A fuse, but as we found from the fuse chart on page 5 fuses don't always work to plan, so I'll give it a try into 4 ohms shortly using a 4 amp fuse for starters.

Anyway, we should be able to work out what it should do...

Maximum "undistorted" output will vary to a degree with load. Early on in this discussion when I was talking about the power supply I commented about how much current was actually required, and if rms watts is rms current times rms volts then how could say a 300VA transformer be used for a 100WPC amplifier?

It might seem like 300VA is 300 AC watts of supply and that should easily translate into 2 x 100 watts, after all there'd be 100 watts to spare? But here I am using a 400VA transformer and 50WPC is pulling the voltage down to exactly what it should be, within 0.7 of a volt.

This is because 400VA divides down by the 48V secondary voltage to 8.33A, but then the rectifier relationship gives DC current as 0.62 x AC, and so 8.33 x 0.62 = 5.16A.

At 50W into 8 ohms the rms voltage is 20V. And 50/20 = 2.5A. There are two channels so that's 5A, so except for the 0.16A, all the current has been used!

Looking at voltage we have 65V on load and I had clipping at around 58V p-p. I have circuit losses of about 5V, so 58 + 5 = 63V, so where is the other 2V?

What about ripple voltage? Reservoir capacitors smooth the supply but current is being drawn all the time, so the bucket is being emptied at the same time as being filled. Making the reservoir a total of 28,200uF (6 rather large 4700uF capacitors) reduces ripple to around 2V, so that's where the other 2V is.

So now we should be able to calculate, even though approximately, what a 4 ohm load will do.

First of all we can double the ripple to 4 volts, because twice the current will be required.

Then we find that the transformer will be overloaded. Yes, a 400VA transformer will be pulled down by us trying to do 100WPC. Food for thought?

Let's say the overload will equal its regulation of 7.5%, so it will be doing just over 44V instead of 48V.

So in DC we have 62.2 volts less the bridge rectifier's 2.2V, equals 60V. And 4 volts of ripple loss making 56V HT. I have 5V of circuit losses so that's now 51V. Divide that by 2 and take the square root = 18V rms.

V squared over 4 gives 81W as the answer.

So how do they make a 50WPC into 8 amp do 90WPC into 4 using a 300VA transformer?

You've heard the saying "it's all done by mirrors"? And here you can take that literally because such designs use current mirrors (and other techniques) which up the voltage gain, so whereas this design has only 80dB open-loop gain, theirs might be 100dB or more (and a lot more transistors and complexity to get there).

There is a lot more negative feedback, and so the onset of clipping is "postponed" by around 0.5dB to 1dB, which is sufficient. The dead giveaway would be the 10kHz and 20kHz THD at 90WPC into 4, but I can't seem to find it...

And then you have 2 x 110WPC into 4 using a 250VA transformer... it's a miracle! Sick


That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Fatmangolf View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 23 Dec 2009
Location: Middlesbrough
Status: Offline
Points: 8960
Post Options Post Options   Thanks (0) Thanks(0)   Quote Fatmangolf Quote  Post ReplyReply Direct Link To This Post Posted: 16 Oct 2018 at 9:54pm
I wonder if mirage may be a better word! This just shows the simple truth of a good sounding amplifier design like yours over something much more complicated built to create impressive stats.

Jon

Open mind and ears whilst owning GSP Genera, Accession M, Accession MC, Elevator EXP, Solo ULDE, Proprius amps, Cusat50 cables, Lautus digital cable, Spatia cables and links, and a Majestic DAC.
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 (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 17 Oct 2018 at 8:17pm
Regarding the question "does the output vary much with load?"

If the load impedance dipped from 8 Ohms to 4 Ohms when the program material constantly produced more than 42 watts reference 8 Ohms, which would mean when playing music you'd be well into hard clipping, and you would never go there (I'm talking about a sane person here), yes the output would vary, it would take out a fuse because it would be exceeding the fuse rating.

Under all other circumstances the output voltage will not vary with load, or in other words, up to 18.3V rms it will be OK if the impedance drops from 8 Ohms to 4 Ohms.

So, into a 4 Ohm load I measured it reaching 84 watts per channel at 0.1% THD+N at 1kHz. So it could be rated as 84WPC into 4 Ohms.

However, we also want to know its performance at 10kHz and 20kHz, and that should decide its rated output (most don't bother with this technicality).

In my opinion 20kHz distortion should equal or better 1% and not exceed it at such extremes, because we are selling an amplifier for the audio frequency range, and that extends to 20kHz.

Into 4 Ohms at 20kHz the output stage beta is dropping so its distortion is rising, and we don't want that causing damage to expensive tweeters, so whilst others might be happy claiming 84 watts, I decided it best to derate it through fuse size, which at least offers some tweeter protection.

And I reckon 75 watts into 4 Ohms 20Hz - 20kHz is quite adequate. The fuses sizes were decided on empirically as 2.5A HT and 3.15A output.

The input sensitivity was noted as 668.8mV, and the mains voltage checked out at 230V (via a large variac). The dummy loads were checked before (cold) and after (v. hot) and found to be 4 Ohms exact. The dummy load being non-inductive means that it is still 4 Ohms at 20kHz. Wire wound dummy loads can vary considerable producing dodgy results.

At 84WPC into 4 ohms both channels driven continuously THD+N at 1kHz: 0.1%

At 75WPC into 4 ohms both channels driven continuously THD+N at 1kHz: 0.09%; at 10kHz: 0.25%; at 20kHz: 0.8%

Intermodulation at 75WPC into 4 ohms both channels driven continuously: 0.4%

Frequency response: 25Hz - 32kHz -0.5dB

S/N: 76dB

Crosstalk: -56dB

HT measured 71V off load and 62.5V on load

Output "mid-point" measured 37.5V off load and 31V on-load (this voltage does not reach the output terminals).

The transformer held up quite well, better than expected.
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 2122232425 345>
  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 0.109 seconds.