[annamarykahn]

My recommendation is what it is, and just because other types of equipment use them doesn't mean I endorse them nor believe those wouldn't sound better without them.

this was just something that i wanted to experiment with ... as a test ... if it ruins the sound, i'll remove it... hey, its only money, lol

[Folsom]

So are you saying that in your opinion it's not a requirement to have a fuse?

Not for any form of protection for other equipment connected to it, no. So long as you're not wiring the hot to the enclosure this is one of the safest amplifiers out there. If it were a transistor driven unit then it would be advisable. 

Technically so long as you isolate the chip tab from the heatsink, there's no requirement to even ground the enclosure because of the complete isolation. But you can do so anyway (I did).

[lacro]

Would you elaborate on this warning a bit more? I am using a 24V SMPS (laptop brick). Are you saying I need to discharge the bricks caps if I disconnect, and reconnect power?? Is it safe to unplug the SMPS from mains, but leave it connected to amp, and then re-plug back into power mains?


 UNDER NO CIRCURMSTANCES connect the PSU board or any other power supply to the amplifier board with a charge in capacitors, or without a switch for say batteries. The spark that can occur can toast the transistors. If you test the PSU then you must discharge the capacitors before connecting the amplifier to it. There’s a place on the PSU board for a discharge resistor (I haven’t defined a size, because it varies for the wide use of the Antipole, 10kohm would be fine for this project)

[Folsom]

Hypex (SMPS) has a good RFI/EMI filter inside. Maybe a Schaffner filter ruin the sound.

What RFI/EMI filter incorporates your design?

Hypex uses much larger filters with the type of parts I use for discrete filters on inputs/on AC. And people have repeatedly reported bad sound from them being plugged into many types of power conditioners, so it wouldn't be a big surprise if people thought those little IEC devices were bad with say some NC500 amps.

[Folsom]

Would you elaborate on this warning a bit more? I am using a 24V SMPS (laptop brick). Are you saying I need to discharge the bricks caps if I disconnect, and reconnect power?? Is it safe to unplug the SMPS from mains, but leave it connected to amp, and then re-plug back into power mains?


 UNDER NO CIRCURMSTANCES connect the PSU board or any other power supply to the amplifier board with a charge in capacitors, or without a switch for say batteries. The spark that can occur can toast the transistors. If you test the PSU then you must discharge the capacitors before connecting the amplifier to it. There’s a place on the PSU board for a discharge resistor (I haven’t defined a size, because it varies for the wide use of the Antipole, 10kohm would be fine for this project)

I would discharge the SMPS's caps prior to connecting it to amplifier, yes. The spike that can happen when an arc occurs can be too much for the transistors. For battery users the solution is using a switch; that may work for you as well. Switches reduce the spike, but I still prefer them on the AC because it's totally safe.

Unplugging and plugging back in SMPS from mains while SMPS is still connected to amplifier is perfectly safe, you could do it all day.

[maty]

You can add a fuse with:

-> http://www2.mouser.com/_/?Keyword=schurter+switch

[maty]

Class D amplifiers are very fast and they have problems with conditioners.

But RFI/EMI one stage filter (without conditioner) is a good idea with ALL amplifiers.

Class A, class AB... are not than fast and more of they can use two stage filters (and Würth 150 kHz ferrites). A fast class AB amplifiers, one stage filter only.

My class AB AV Marantz SR4500 and AV Yamaha RX-V2700: two Schaffner cables and one 8mm Würth 150 kHz ferrite.

My class A ifi iCAN (headphones amp) is very fast, with bandwith up 500 kHz. One Schaffner cable and one Würth 150 kHz 12.5 mm (two rounds) ferrite.

-> http://aficionados.foroactivo.com/t197p210-usb-isolator-ifi-iusb-vs-teradak-u9va-teralink-adum4160-filtros-schaffner-lampizator-ac-filter-diy-esa-silk-estabilizador-regulador-de-tension-voltaje-isolation-balanced-transformer-dc-blocker-blocking-pc-silencioso-en-aussar-variac#16121

To English -> https://translate.google.com/translate?sl=es&tl=en&js=y&prev=_t&hl=es&ie=UTF-8&u=http%3A%2F%2Faficionados.foroactivo.com%2Ft197p210-usb-isolator-ifi-iusb-vs-teradak-u9va-teralink-adum4160-filtros-schaffner-lampizator-ac-filter-diy-esa-silk-estabilizador-regulador-de-tension-voltaje-isolation-balanced-transformer-dc-blocker-blocking-pc-silencioso-en-aussar-variac%2316121&edit-text=

My second system.

RFI/EMI Schaffner filters + Würth 150 kHz ferrites (power and RCA cables) and DC & Ripple Blocker to ALL SYSTEM.

I have:

1 Shaffner + 1 Würth ferrite => Power Strip => DC & Ripple Blocker (4 outputs, in star) =>

=> 1 Schaffner -> AV Marantz SR 4500

=> ifi iCAN

=> 1 Schaffner -> PC (with Bequiet! L8 400 PSU: 5mV/10mV ripple in 5V – USB)

1 Schaffner + 1 Würth 150 kHz -> old 17″ Hitachi CRT monitor (1998), with very good image!

PC => 1 Würth 150 kHz (two rounds) in cheap USB cable -> ODAC =>

ODAC -> RCA cable with 2 Würth 150 kHz ferrites -> ifi iCAN

ODAC -> RCA cable with 2 Würth 150 kHz ferrites -> AV Marantz SR4500

And now, in my KEF Q100 speakers, a Würth 150 kHz ferrite ONLY in woofer cable.

********** **********

By the way, the ifi iCAN does not have a fuse!!! And the others ifi amps and DACs not   

[annamarykahn]

Technically so long as you isolate the chip tab from the heatsink, there's no requirement to even ground the enclosure because of the complete isolation. But you can do so anyway (I did).

i was going to ask about the reason for isolating the chip tab from the heatsink ... the drok tda7297 that i got from amazon isn't ... can someone elaborate on the reasons for isolation? it also doesn't have heat sink compound, which i didn't like, so i added my own, which was when i discovered that it wasn't isolated

thanks!

[Folsom]

maty, there's good ideas and what works, too. Honestly I feel like you've made a topic within a topic, it's a bit obtrusive.

I've got a PCB board that's never been printed that's for offering an input filter that isn't so cheap as the IEC version. If there was sufficient interest for a buy in to get them printed I'd do that.

As far as filtering the Antipole features AC filtration that's very effective. You'd think because the transformer provides such low impedance that the CMC and capacitors would do a lot, but that's not the case. The caps work better because of the CMC in place, first, and second if you swap between CMC/no-CMC by soldering directly to the CMC legs you can quickly and easily hear a very audible difference.

The boards are also based on copper pours (plains). The amp itself has better PSRR from the KMR section that provides low impedance, and the OSCON's in parallel drop their impedance to almost nothing.

The transformer I've been recommending also is good at rejecting noise since it's a dual bobbin.

There's quite a bit of work for noise done on this amplifier. That's a huge focus with anything I do.

[maty]

The main is increasingly dirty by the cheap switched transformer connecting.

Anyway, everyone decides for himself that enough information has been provided.

[Folsom]

i was going to ask about the reason for isolating the chip tab from the heatsink ... the drok tda7297 that i got from amazon isn't ... can someone elaborate on the reasons for isolation? it also doesn't have heat sink compound, which i didn't like, so i added my own, which was when i discovered that it wasn't isolated

thanks!

With my amplifier's PCB board it provides a very low impedance path for RF that can go through the chip. The chinese ones don't have this so the RF doesn't go through the chip nearly as much. Also I use bigger heatsinks (and recommend) on top of a metal enclosure so it all becomes a potential antenna. The enclosure and heatsink can either act like a shield or an antenna depending on the paths you allow.

The chinese versions basically lower their performance greatly by designing the way they did, which avoids some RF, but doesn't take anywhere near full advantage of the chip. They sound nice, and would be great for a TV, but as far as being truly high end their high impedance, high noise, often bad grounding paths, etc, aren't going to get them there. For $6 they're a bargain so long as you can resolder all the connections and hope the pot didn't break in shipping. They're especially good if modified. But they aren't high end. The chips aren't even up to standards.

[Folsom]

The main is increasingly dirty by the cheap switched transformer connecting.

Can we get a better translation on that Maty?

I'd say you should get some boards, but you'll need some more efficient speakers

[maty]

The guilty:

I apologise for my poor English.

[Folsom]

I came up with a layout I like more than the demo, thought I'd share with you all considering enclosures. You'll notice I didn't include a switch because location for that is your choice. I kind of like them in back,

[annamarykahn]

i'm in the process of building/planning an enclosure ... i was thinking of using wood ... is that a bad idea?

[G Georgopoulos]

The main is increasingly dirty by the cheap switched transformer connecting.

Anyway, everyone decides for himself that enough information has been provided.

Hi maty,nice avatar Mate!

it's not mains noise but rather smps operating frequency,you can get rid of it by an C L C type pi filter folowing the smps plugpack, it's that simple...

all the best Maty...

[Folsom]

i'm in the process of building/planning an enclosure ... i was thinking of using wood ... is that a bad idea?

It provides no shielding of any kind. Performance won't be as good. It may look nice if you'd good with wood however.

The back panel where you mount everything is very challenging on wood, it's much easier with a metal plate of sorts, at least.

Also it's best to have some distance from transformer and amplifier board.

[BRN]

I came up with a layout I like more than the demo, thought I'd share with you all considering enclosures. You'll notice I didn't include a switch because location for that is your choice. I kind of like them in back,

I like this layout. It is very clean, providing separation between the power supply and amp. I just need to make sure that there is room between the amp and the speaker terminals.

[Folsom]

Standoff height should help you do that.

[devinkato]

Any reason why there is so much extra space in that layout, especially on the amp side?  Or is that a standard chassis size?