Acurus A250 and starground

[KNielsen]

Hey Ak

I'm restoring my Acurus A250 and at the same time making some mods, this includes new starground curciut.
I'm splitting input and feedbackground (now they share same ground at input), but what about D3/D6...??? I parallel a 4,7uf/470nf cap over these for lower noise, but should I also split these diodes from inputground to a better location for a better signal separation in the curciut it self?

Hope my poor english is readable

Regards

 

[KNielsen]

My A250 rebuild so far...

Dale RN60/70-series resistors.
Mills 5w at the emitters.
Nichicon muse 470uf at C7.
Panasonic Ecwf (a) to replase all values between 0.1-2.2uf.
Charcroft Audio Silver Mica.
Jantzen Alumen 1uf/Vishay Atom 50uf parallel at DC rail.
Kemet 13000uf 100v (8 of them) at Psu.
New bridge, 2 Hexfred 1200v 55a each.

Edit: this is not how the caps at the PSU is going to be placed. Just the pic i had at hand

Regards

 

[mondialfan]

I would be careful with separating a bunch of the grounds out from the components on the board. If you're going with separate rectifiers and capacitor banks for each channel then you'll have a central ground point on each channels ground bus bar with a chassis ground wire going back to the central chassis star ground. I'd probably leave the board grounds alone except for adding the necessary ground path for the new post fuse DC rail capacitors.

Also, avoid the urge to go too much larger (physically larger) on the input coupling capacitor C1. Many times people remove this cap from the board and install a much larger boutique film cap inline here and it adds more noise to the circuit because that large cap is also a big antenna. If you put the amp on a signal or distortion analyzer many times you'll see a large increase in noise after installing a large film coupling cap. I've even seen some amps with oscillation issues due to a massive film cap installed in the input. If you're going to do this I wouldn't attempt it without an o-scope and signal analyzer to look at the amp before and after installing the large coupling cap. Then you can attempt to relocate the cap if it introduces noise.

For the 470uF C7 feedback down capacitor you're shifting the low frequency cut-off point down quite a bit. With the stock 220uF C7 I measured -.25 to -.3dB @ 10Hz on the last A200 (identical circuit) I had on my bench. That's pretty good low frequency response...

 

[KNielsen]

I would be careful with separating a bunch of the grounds out from the components on the board. If you're going with separate rectifiers and capacitor banks for each channel then you'll have a central ground point on each channels ground bus bar with a chassis ground wire going back to the central chassis star ground. I'd probably leave the board grounds alone except for adding the necessary ground path for the new post fuse DC rail capacitors.

Also, avoid the urge to go too much larger (physically larger) on the input coupling capacitor C1. Many times people remove this cap from the board and install a much larger boutique film cap inline here and it adds more noise to the circuit because that large cap is also a big antenna. If you put the amp on a signal or distortion analyzer many times you'll see a large increase in noise after installing a large film coupling cap. I've even seen some amps with oscillation issues due to a massive film cap installed in the input. If you're going to do this I wouldn't attempt it without an o-scope and signal analyzer to look at the amp before and after installing the large coupling cap. Then you can attempt to relocate the cap if it introduces noise.

For the 470uF C7 feedback down capacitor you're shifting the low frequency cut-off point down quite a bit. With the stock 220uF C7 I measured -.25 to -.3dB @ 10Hz on the last A200 (identical circuit) I had on my bench. That's pretty good low frequency response...

Thanks for reply.
Actualy I was planning for a 10uf film, but maby a Nichicon Muse would be better

I'll keeper the capacitor bank as original, but with 104000uf, should be enough to avoid crosstalk. The two bridges is coupled as one for the better amps. - A software is added.

Im unsure about the caps for the DC rail (taken from Aragon 4004).. Now I have a 50uf polar and a Jantzen Audio Alumen 100v 1uf filmcap in parallel. But is 100v enough on the filmtype?? I have thought about Panasonic ECWF (a) for this location instead.?

Regards

 

[mondialfan]

Actualy I was planning for a 10uf film, but maby a Nichicon Muse would be better

I'll keeper the psu as original, but with 104000uf, should be enough to avoid crosstalk. The two bridges is coupled as one for the better amps. - A software is added.

What caps should I use for the DC rail (taken from Aragon 4004).. Now I have a 50uf polar and a Jantzen Audio Alumen 100v 1uf filmcap in parallel. But I'm unsure about my choice on the film, is the voltage ok?? I have thought about Panasonic ECWF (a) for this location instead.?

Regards

If you have 8 Kemets and two bridges I would run the two bridges separately instead of bridged. One bridge and 4 Kemets per channel. This will break the supply up to what I call pseudo dual-mono. The two channels use the same transformer windings but everything after that is channel dependent. This should help to minimize channel interaction.

Definitely leave out the big 10uF film and go with the Nichicon Muse and an ECWF on the input coupling cap. You should be able to squeeze a .33 or .47uF ECWF in a similar physical size into the C1 location.

You can use the 50uF/100V but I usually go with a 100uF/100v polar on the rails, ECWF films are good. There is a lot of open area on the board so I usually opt for a radial cap and drill a couple of small holes to install it then a small dab of glue to hold it in place and then P2P wire on the back side of the board. In this manner it looks like the factory installed it.

 

[KNielsen]

Thanks helps alot.

How about C12. Should I leave the value as it is? I've changed this to ECWF 0.1uf, but would a higher uf be better?

 

[mondialfan]

Thanks helps alot.

How about C12. Should I leave the value as it is? I've changed this to ECWF 0.1uf, but would a higher uf be better?

I would leave C12 as is.

Here's a photo from a recent A200 upgrade which shows the board mounted post fuse DC rail e-caps (the brown ones on either end of the board). The bypass films are ECWF's and are mounted in the center of the board between the rail fuses. All wiring is P2P and located behind the board using the original ground planes.

BTW, this amp put out .002% - .007% THD+N from 2W - 250wpc @ 8ohms, 1kHz & about 260 WPC @ the factory rated .06% distortion.

 

[KNielsen]

I would leave C12 as is.

Here's a photo from a recent A200 upgrade which shows the board mounted post fuse DC rail e-caps (the brown ones on either end of the board). The bypass films are ECWF's and are mounted in the center of the board between the rail fuses. All wiring is P2P and located behind the board using the original ground planes.

View attachment 1146347

Looks nice and very professional.

Have you ever separated the inputground from the negative speaker plug?
If I go all the way with starground, a separation of input, output and chassic would be the way to go.

And one final question. Witch resister are in parallel on you psu-caps? Ohm/watt??

Thanks for your time

 

[mondialfan]

The back to back diodes and resistor should provide the needed isolation from the speaker ground. The later Acurus amps put these components on the boards instead of just inline. It seemed that Mondial realized the back to back diodes were needed and added them after the fact on the Gen 1 amps, thats why they're inline and under heatshrink.

The later amps also had separate wires from the board grounds. For the front end circuitry the wire goes back to star ground directly and for the output it goes to the neg speaker binding post and then to star ground. So there are two grounds per board, both tied back to the star ground.

For the resistors I used Panasonic ERG 3W/5.6k ohm. I usually decide the value based on the rail voltage (~84VDC for the A200 shown, but I have 125.3VAC from the wall here) and I try to keep the dissipation closer to 1/3 of rating for longevity. I know a lot of folks say up to 50% is good but I like to be more conservative than that.

 

[KNielsen]

Thanks Mondialfan

I'll post some pics here as the work progress

Regards

 

[KNielsen]

Any suggestions on the Zener diodes D3/D6.

Have anyone tried to parallel these with a cap for lower noise?

Regards

 

[mondialfan]

I have done this on a number of amps and preamps, not just Acurus. It can have a positive effect especially if the zener isn't being operated much above the avalanche and is putting excessive noise on the reference. However you have to be careful as on some amps I've experienced a "Thump" through the speakers on power up, usually when I use a value large enough that it slows the charging time. Due to this if I'm going to install this cap I usually limit it to a film cap of a low value, i.e. < 0.47uF.

 

[KNielsen]

I have done this on a number of amps and preamps, not just Acurus. It can have a positive effect especially if the zener isn't being operated much above the avalanche and is putting excessive noise on the reference. However you have to be careful as on some amps I've experienced a "Thump" through the speakers on power up, usually when I use a value large enough that it slows the charging time. Due to this if I'm going to install this cap I usually limit it to a film cap of a low value, i.e. < 0.47uF.

Thanks. I cancel the 4.7uf I was planing.
Can I use a ECWF(a) .22~.47uf on these diodes of should I look for another brand..?

 

[mondialfan]

ECWF should be fine. I like them because the leads are longer which makes them easier to position. A lot easier than say a Wima MKP with 4 or 5mm leads.

 

[KNielsen]

Mondialfan: I have to order some parts to complete the circuit, and i'm wondering witch mica insulation to go whit between the transistor and the cooling profile..?? I'm ordering from mouser.

Regards

 

[mondialfan]

I don't think I've seen the correct size mica at Mouser. I bought a large pack of mica insulators (like hundreds of them) on Ebay years ago. I'm still using those.

 

[KNielsen]

I would leave C12 as is.

Here's a photo from a recent A200 upgrade which shows the board mounted post fuse DC rail e-caps (the brown ones on either end of the board). The bypass films are ECWF's and are mounted in the center of the board between the rail fuses. All wiring is P2P and located behind the board using the original ground planes.

BTW, this amp put out .002% - .007% THD+N from 2W - 250wpc @ 8ohms, 1kHz & about 260 WPC @ the factory rated .06% distortion.

View attachment 1146347

The project has been on the shelf for some time, its been a good summer in Denmark

Anyway I'm back to continue..
- The P2P wiring on the post fuse caps, are you returning them to the circuit ground output (speaker GND on the board)???

Thanks

 

[mondialfan]

Yes, wire them back to the speaker ground on the board.

 

[mondialfan]

The schematic you're working with is the worst of the two I've seen floating around on the net. If you'd like a better schematic to work with I recently took the better quality A250 schematic and cleaned it up further and made needed changes to it. I attached a copy. I have a much higher res .bmp of it if you'd like to contact me via PM I can send it to you.

 

[KNielsen]

I'm going to do some metalwork on the cabinet. Moving the AC-input form left to center on the chassis, making room for the PSU caps come as close to the boards as possible in both sides. The original AC hole is welded and closed, then painted..

Haven't removed the old clambs yet