Bridgeclone - Rowland Concentraclone

Status
Not open for further replies.
Jump to Latest
ThorstenL

ThorstenL

Previously: Kuei Yang Wang
Joined 2002
linkedin.com/in/thorsten-loesch-studio-raumklangg/
#1
Koinichiwa,

I just came across some hi-rez internal shots from the new Cncentra. It may be of interest to those seeking to "clown around" with "Bridges over the River Kwai" and the like....

First the overview:

An externally hosted image should be here but it was not working when we last tested it.


As can be seen, unlike the other Rowland Amps that use switch mode supplies, the concentra still does it the oldfashioned way...

The inputs are switched by little, very familiar looking relais, goldflash over copper contact IIRC and then you get a nice Jensen Input Transformer:

An externally hosted image should be here but it was not working when we last tested it.


There is a lot of jiggery pokery and to those who already know, the Concentra uses the CS3310 as volume control.

The following picture shows parts of the PSU, pretty basic, single bridge with ceramic snubber caps, 4 pcs Nichicon PSU Cap's, I'd guess from the looks 10,000uF/63V each:

An externally hosted image should be here but it was not working when we last tested it.


Now onwards to one of the six Nat Semi LM38XX (we know they are 3886) output chips in bridge/parallel:

An externally hosted image should be here but it was not working when we last tested it.


No sign of a servo. Dale 0.2R 1% Output Resistor. All else SMD, including the two local decoupling caps which are by the looks X7R or possibly even Z/Y5U Multilayer Ceramics. All the other SMD resistors appear to be tight tolerance (guess) and there is no manual offset correction either. The paucity of passive components suggests BTW inverting mode....

Sayonara
 
#5
Very nice clamping bar holding the IC's onto the heatsink. It puts the pressure right on the part of the tab that has the most heat. Way better than using the bolt hole that pulls the IC down unevenly. Good method for TO-220's and TO-247's too, especially if you are working them hard.
 
#6
Circlotron said:
Very nice clamping bar holding the IC's onto the heatsink. It puts the pressure right on the part of the tab that has the most heat. Way better than using the bolt hole that pulls the IC down unevenly. Good method for TO-220's and TO-247's too, especially if you are working them hard.
Click to expand...

Don't forget the nice SS fasteners, the only screws I use. 😉
 
#12
mothman said:
A beautiful amp indeed and if I had the extra $7000 I'd buy one!
I'd settle for a dozen of the chip insulator pads though at this point.
Anybody know where a fit pad like those can be had?
Click to expand...


If you run accross them I would love to know where, I've been looking for them too! That said, there is a universal part at Digi-key, part number BER190-ND that looks like it will work.

Scott
 
#14
mothman....I buy sil-pad by the sheet (12x12) and cut out what I want...its a tenth of the price that way....I know there are better insulaters out there...but I dont cut my corners on cooling and good thermal compound works for me


DIRT®
 
#16
Most big guys have the sheets but since we are talking the states why does nobody look at for instance Digi-Key? Seems to be a fairly big supplier? They seem to have plenty but i don't know the company very well and not their catalogue either. Try any of the following most obvious potential candidates, all Multi-Watt as it seems, from their on-line catalogue:

BER190-ND
GAPPADHC100020MIL-ND
K4AC-124-ND

Some info :
http://dkc3.digikey.com/PDF/T032/0480.pdf
 
#17
As regards clamping down the chip with something like those metal bars, I presume that the chips do have a breaking point and I wondered if the bars have to be carefully torqued down?

And I can just envisage a thread here about 'the best sounding torque setting' :bawling:

Personally, I would prefer something springy with a bit of give to be on the safe side!
 
#18
I saw some internal pictures some time ago of the Linn 2250 power amp, on a spanish magazine.
It uses what seams to me as 3 x TDA7294 per channel :dead: .
The pictures were not very clear...
I wonder what topology do they use.
3 chips per channel...:scratch:
3 parallel?
They claim 115w RMS / 8 ohms, 230w RMS / 4 ohms.
 
#19
miguel2 said:
The signal goes directly from the jensens to the vol control and then to the output chips? No other gain/buffer ICs in between?

Miguel
Click to expand...


The CS3310 is a buffer/gain stage too.
Check here:
http://www.cirrus.com/en/products/pro/detail/P2.html

It says:
"The CS3310 includes an array of well-matched resistors and a low noise active output stage that is capable of driving a 600 W load. A total adjustable range of 127 dB, in 0.5 dB steps, is achieved through 95.5 dB of attenuation and 31.5 dB of gain"
 
Status
Not open for further replies.
Top Bottom