Copyright © 1998
Cadman Enterprises Ltd
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 | Page <1 103104105106107 345> |
Author | ||
Graham Slee
Admin Group Retired Joined: 11 Jan 2008 Location: South Yorkshire Status: Offline Points: 16298 |
Post Options
Thanks(1)
|
|
As can be seen from the FFT below, the 38k and 46k noise are still there, but the 20kHz noise is gone (but as it was still daylight, the grid-tied inverters, of which there are eight within 500 feet, would still be putting 20kHz on the mains). The HT filter responsible for attenuating HF noise will be discussed in another post. The 38 to 46 has only one thing in common - the laptop - and as the laptop is running the AP software, It cannot be eliminated by switching it off. The 1k resistor test proved one thing - that it is nothing to do with the amp! However, there is a spike at 58kHz, which, although it is at -93dB, might be suggestive of a parasitic trying to establish itself, but It is doubtful it will get anywhere. Above that, the noise is between 8uV and 24uV, and similar in frequency to the 1k resistor noise FFT. 100kHz is about -3dB below lower frequencies, roughly the amp's turnover, and then it can be seen to roll-off at 6dB per octave to 400kHz. Before the instrument's cut-off frequency, which is 500kHz with two channels running, there is a surge in high-frequency noise. When subtracted from the 1k resistor noise (which extends out to 1MHz, because it can work with only one channel), it will be noted that the amp's HF is still falling off at the correct rate. The LF noise is -80dB, and below from 100Hz, and increasing to -76dB below at around 40Hz, as it approaches the 1/f noise phenomena. The rms noise measured -76dB, and -84dB "A" weighted. Although it can be seen half of the noise in the audio spectrum is -100dB, the noise reading computes the sum of the averages. Edited by Graham Slee - 05 Jul 2020 at 3:42pm |
||
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
|
||
gwebster
Senior Member Joined: 04 Oct 2014 Location: Liss, UK Status: Offline Points: 175 |
Post Options
Thanks(0)
|
|
Not sure the FM and DAB bands are the issue here, they're up in the 100MHz and 200MHz zones. Must be something else causing the kHz stuff
|
||
Graham Slee
Admin Group Retired Joined: 11 Jan 2008 Location: South Yorkshire Status: Offline Points: 16298 |
Post Options
Thanks(0)
|
|
Brain outage! Senility setting in? I will leave your comment but will edit out my garbage!
Edited by Graham Slee - 05 Jul 2020 at 3:44pm |
||
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
|
||
BackinBlack
Senior Member Joined: 05 Feb 2012 Location: Hinton, N'hants Status: Offline Points: 2020 |
Post Options
Thanks(0)
|
|
As ever this is all fascinating and quite enlightening.
From your observations of interference it would seem that our mains supplies are perhaps beyond redemption. As the number of devices increase, not just low power wall wart SMPS but also EV battery chargers, we shall see ever more noise on our power networks.
The apparent lack of suitable conventional linear power supply capacitors that are able to adequately suppress this noise does lead to the question as to whether simple linear power supplies are worth pursuing. Perhaps we shall have to move toward either regulated battery supplies (expensive or impractical for power amps) or join the SMPS brigade? Then of course there's the ever growing digital communications devices all around us to contend with. I don't see any other manufacturers claiming to address (and overcome) the curse of EMI interference without using power regenerator or conditioning devices. It's probably consigned to the "too difficult" box and ignored. |
||
Just listen, if it sounds good to you, enjoy it.
|
||
Graham Slee
Admin Group Retired Joined: 11 Jan 2008 Location: South Yorkshire Status: Offline Points: 16298 |
Post Options
Thanks(0)
|
|
Test equipment is not immune to noise, and it will display it wherever it comes from.
In the 1k resistor test, the 1 k resistor was placed across the AP input, and was nothing to do with the amplifier. I guess it could be difficult for a small 10mm long resistor and short wire ends, to pick-up airborne 20kHz transmission? If so, then the 20kHz tone must be getting through the AP's power supply. To be honest, I knew nothing of grid-tied inverters causing 20kHz dirt on the mains, but by researching grid-tied inverters, it wasn't being denied. There are eight rooftop solar panel installations within 50 to 500 feet of my workshop, which can be seen on Google maps. My opinion is that allowing voltages on the mains other than those at the generator frequency (50Hz here) is not supposed to happen. Maybe I'm wrong? I have a Roxburgh RIR0422H mains filter on the amp input. However, it is rated at 4 amps, and, according to Morgan Jones, I might have made a mistake here. The peak current drawn by the amplifier is 7.4 amps (ac), and that equates to 1.5 amps on the primary side. Morgan Jones illustrates that ripple can be 10 times the load current, and therefore the filter should be rated 15 amps. So, even running with quiescent current most of the 24/7, and only at 1 watt out, when listening, which allegedly takes around 2 - 3 amps from the mains, it must have burnt-out on peak measurements! It is odd however, that the 2.5 amp mains fuse remained intact. The transformer is low saturation, and has an interwinding screen. I don't think the customer needs to go to the added expense of an isolating transformer in the mains supply! But let us assume this 20kHz is getting through, and being rectified. The electrolytic smoothing capacitors aren't good for 20kHz. It also seems like the 20kHz is getting through the AP supply! What we want is to rectify and smooth the mains, but the solar panel installers add the 20kHz, legally? Well, if it happens here, it could happen anywhere. |
||
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
|
||
Graham Slee
Admin Group Retired Joined: 11 Jan 2008 Location: South Yorkshire Status: Offline Points: 16298 |
Post Options
Thanks(0)
|
|
Solar Panel white papers showing the 20kHz switching frequency of converters.
https://core.ac.uk/download/pdf/74235679.pdf https://pureportal.strath.ac.uk/files-asset/64261014/Zakzouk_etal_JESTPE2016_PV_single_phase_grid_connected_converter.pdfYou https://trace.tennessee.edu/cgi/viewcontent.cgi?referer=https://www.google.com/&httpsredir=1&article=2473&context=utk_gradthes Make of it what you will. |
||
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
|
||
Graham Slee
Admin Group Retired Joined: 11 Jan 2008 Location: South Yorkshire Status: Offline Points: 16298 |
Post Options
Thanks(0)
|
|
If it's reliability you're after, this amplifier has it in piles!
Provided you don't do anything stupid, this amplifier will accept hundreds of modifications, without going into self destruct. It will put up with a dead short on one channel, while "complaining like hell" on the other channel! And it's because the output transistors are tough, really tough, like overcooked beef steak tough. Morgan Jones explains this in the image below. High-frequency power transistors are achieved through epitaxial means, a form of compounding, which in-effect means many paralleled smaller transistor bases. This is supposedly better than homotaxial because it enables the transistor to have increased beta up to a higher frequency. Except that, if we take a long look back into history, we find the better power amps used a different version of the then ubiquitous 2N3055. Often these were known by a different RCA number, and other manufacturers added the suffix H. Although Morgan Jones makes a valid point regarding the slowness, and the H suffix devices were considerably slower, they also demanded higher base current at the frequencies where parasitics would otherwise exist. These parasitics being caused by negative impedance. The H suffix devices could also withstand short circuits for a longer duration, and this could be due to their thicker base. They were also often utilised without emitter resistors, and this suggests to me that their capacitance might have offset oscillatory problems. As I said, the H suffix versions tended to be given a different part number, and those are found in some of the highly acclaimed amplifiers of the day. |
||
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
|
||
Post Reply | Page <1 103104105106107 345> |
Tweet
|
Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You can vote in polls in this forum |