Building from a pair of MA-9S2 boards, first project

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I just bought this pair of boards which implement a clone of the Marantz MA-9S2. They're all assembled and look more or less like the picture on the vendor's page at Wholesale Product Snapshot Product name is Assembled 150W+150W HIFI audio power amplifier clone Marantz MA-9S2 AMP C5200/A1943 free shipping. Each board is one channel.

Although I think I understand pretty much what to do next, I am getting the feeling I also might be in over my head as this is my first amplifier project, close to the first thing I've done with analog electronics, and the first project which requires that I build a power supply from parts. I am really just starting to learn and so there are gigantic gaps in my knowledge. I'm hoping that finishing this project will end up allowing me to banish at least some of the major gaps. I'm confident I can eventually puzzle this thing out, but would sure appreciate some help!

I understand that each of these boards wants a nice toroidal power supply to give it around 25v DC, so I figure I will get this toroid: 300VA Toroidal Transformers and then hook each output up through a bridge rectifier like this: Wholesale Product Snapshot Product name is KBPC5010 Bridge Rectifiers DIODE RECT BRIDGE 50A 1000V, followed by a capacitor or two for smoothing--that last part I'm less sure of how to do right since I am starting to think there are probably some important subtleties.

Does it sound to y'all as if I am more or less on the right track? I also still need to figure out where to attach things on these boards. The places that look like attachment points on the board, followed by what I think they're for, are:

GND: to be attached to the earth from the wall power

-VSS: to be attached to the negative side of the DC power

+VCC: this must be the positive side of the DC power

a pair, one of which is labeled INPUT+ and the other is unlabeled: this must be the audio input


OUT: One speaker wire must get attached here, but where's the other wire go? Does it get hooked to the same GND?


Okay, that's where I'm at now. I will probably end up searching down some books on power supply design next, and order the toroid as above unless with your help I figure out that that's not the right one.

Oh yeah, one more thing: I am also going to need to figure out how to add a volume adjustment knob/knobs. From my initial reading it looks like a voltage divider circuit might be the thing, but I'm just starting to read about that and would very much appreciate advice or pointers.

I feel like there is probably a book which I should read to tell me a lot of this stuff. If there is one that folks here would recommend, I would love to find out what it is.

Whew!
 
Hi,

Looking for an amplifier for my daughter, I came across the same amplifier. Compared to the original MA-9S2 the schematic appears to be copied with some alterations regarding Power-stage (one transistor pair less), Bias setting and the protection circuit, which has been omitted. on the other hand a DC-servo has been added (does nothing if the power-stage fails).
The original MA-9S2 uses the big (Toshiba?) power transistors, which are of higher spec (power/current) than the 5200/1943 (only 2 pairs) used in the clone. My idea would be to modify the clone's power-stage and add at least one pair of 5200/1943.

Power supply:
the ad. I've found, states that the amp requires +/- 30V to 63V (max. rating of capacitors). So a 2x25 Volts will answer to that requirement. On the top level, the load regulation (5 to 7%) of the transformer MUST be regarded. Thus the max. voltage would be 2x40V.... or change the caps on the board to 80V or 100V types. For Vss and Vcc: use enough smoothing capacity. At least 2x15.000uF (and an inrush limiter). And preferably a double rectifier.

Grounding:
Try using STAR-grounding. Also the "-" of the speakers ought to be connected there. Be sure the leads to the star-ground are well connected!
 
Just one note:

That is not a clone of the Marantz MA9, and absolutely not any relation with that! Marantz MA9 is more sphisticated, fully balanced from input to output.

Sajti
Hi Sajti,

On what information did you come to this conclusion? As far as I've compared the schematics of the original MA-9S2 and this so called clone, I observed the following:
- The input circuits of both amps use exactly the same layout (1 on 1) except for the values of resistors and the transistors used. This is what I call cloning.
- The bias circuit is different in both amps, however performs the same functionality. This is more a way to keep the Lawyers from your door, by not exactly transposing the architecture of the MA-9S2.
- The power section is in architecture an exact copy of the MA-9S2, scaled down and with less powerful transistors.

In my view this amp is absolutely a clone. Sure the original MA-9S2 excels in building quality, better and matched components and better specs. But the clone is almost for free (less than $50,-)! But that is what cloning for DIY is usually about.

Imho it is quit easy to upgrade such a clone to a level close to the origional. buy 3 pairs of well matched Sanken 2922/1295 (yes, I've looked them up in the meantime) with the proper drivers, change all transistors and resistor values back to original... then you will be closer to the original at a fraction of the cost.

Regards,
Edwin
 
Hi,

the most important difference is, that Marantz use separated voltage amplifier, and power buffer stage to eliminate the back EMF distortion as much as possible.
So MA9 (and even SM11, and PM11) has two individually feedbacked stages. The output stage has a gain about 10-12dB only. The mentioned clone has the gain of 30.

The MA9 is fully balanced from input to the output. The clone is not balanced.

The Marantz has triple darlington output, the clone has double only. (This is not downscale, as the small Maratz amplifiers such as 7001 still use triple darlington)

MA9 use no DC servo, the clone use it.

So this is not MA9 clone. I heard one of this clone, and I heard MA9 too. Believe me, the original is a different world.

Sajti
 
Last edited:
Hi,

the most important difference is, that Marantz use separated voltage amplifier, and power buffer stage to eliminate the back EMF distortion as much as possible.
So MA9 (and even SM11, and PM11) has two individually feedbacked stages. The output stage has a gain about 10-12dB only. The mentioned clone has the gain of 30.

The MA9 is fully balanced from input to the output. The clone is not balanced.

The Marantz has triple darlington output, the clone has double only. (This is not downscale, as the small Maratz amplifiers such as 7001 still use triple darlington)

MA9 use no DC servo, the clone use it.

So this is not MA9 clone. I heard one of this clone, and I heard MA9 too. Believe me, the original is a different world.

Sajti
Hi Sajti,

I do not disagree with you that the original is a world apart from the clone. That's what I wrote, too.

Indeed the overall layout of the original amplifier is different from the clone. The clone combines the P501 and P601 board in terms of amplification, introducing sound degradation. The clone just resembles the P601 board.
And you are right: if the clone also consisted of a copy of the P501 board, it would be a different ball-game... also in court, for the guy selling the clone would have been sued for copyright infringement.

In the original however, one of the HDAM's performs the same function as the opamp DC-servo in the clone....

As far as the use of darlington's is concerned: I only see one in the original amp, which is in the bias circuit. the power stage uses Epitaxial planar transistors (Sanken).

Regards
 
Hi,

in the original amplifier the 2SC3421/2SA1358+2SC4883/2SA1859+2SC2292/2SA1612 form a triple darlington on the output. In the clone the 2SC5171/2SA1930+2SC5200/2SA1943 form a double darlingoton only. This is what I mentioned, not the bias circuit.

Sajti
 
Member
Joined 2010
Paid Member
Building a dual rail power supply from a transformer, bridge rectifier or two and two or more electrolytic caps is fairly straightforward and there are already literally thousands of circuit diagrams on the net that are basically the same. Here, it would be better to understand what the power supply should do and point out what the extra components are specifically for, before asking, "Hey guys, is this right?"

Have a careful read of this article on power supplies from a huge site that treats just about everything and every possible topic for linear audio amplifiers from raw beginner to serious DIY: Linear Power Supply Design
 
Thanks! I think that my predicament is pretty much precisely because there is so much information and discussion available. The design I'm working on attempts to be my synthesis based on what I found which seemed relevant, though I am sufficiently a novice that I would love it were to someone take a look and point out obvious deficiencies or elements which may have arisen from my misunderstandings.

I'll go read that article for sure.
 
Member
Joined 2010
Paid Member
OK, if you are uncertain what to do now but you've decided you need to start with a basic power supply built from parts you source individually, as your schematic suggests, you need to decide a few things like the numbers that will set the limits for your end result.

First though, if you check your schematic against other dual rail supplies, you'll notice you have used further split supplies for each rail, so much of what you have drawn is it is unnecessary. This circuit below isn't exactly what I mean but it's close enough to illustrate the elements and all you need for a basic dual bridge, dual rail supply.
An externally hosted image should be here but it was not working when we last tested it.



A 2 x 25V AC transformer means about 35V DC power rails and that limits the amplifier to perhaps 50W/8R output. Going a little higher to a 2 x 35VAC transformer, gives about 50VDC and 100W/8R. No point going much higher unless your speakers are thirsty monsters and your room is large. A rule of thumb for a single toroidal transformer rating in VA, is 4 X the nominal 8R output power per stereo channel. So if you expect 100W/8R and 150W 4R, you need approximately a 400VA toroid. You could use less, like a standard 300VA transformer, assuming you don't use the amplifier at maximum power levels for more than transient signal peaks. Any diode bridges from 10-50A and 200PIV rated will be fine there. The smoothing caps should be 10,000 - 15,000 uF 63V and the bleed resistors larger, at about 2k 2W, IMV.
 
The sellers say 50-100mA per output transistor ( 10 - 40mV across any 0.22R emitter resistor ). Seems a bit high to me - min distortion is usually 10-15mA per device for class AB. I've built up a couple of these pcbs and will eventually set them up for min thd with my spectrum analyzer. The circuit of the "clone" amps, as mentioned previously, is used by Marantz simply as a low gain output buffer, with voltage amplification provide elsewhere. All of the open loop gain of the "clone" is provided by the two cascode transistors and is necessarily quite low when compared with the open loop gain of, say, a Self blameless amplifier. The upshot of this is higher distortion than that produced by a Self blameless. A possible solution to this problem is to set the gain of the "clone" to 5 - 10 and precede it with a decent opamp ( eg LME49710 ), included within the overall feedback loop to give a final gain of 20 - 30. I'll probably try this sometime.
 
Member
Joined 2010
Paid Member
I assume the seller really means to say "50-100mA per complementary pair" of output transistors. The clone is only a 2-pair EF2 design, unlike the original design. It probably comes with copies of the original transistors or the price would be inflated to NOS level.

50 mA/pair could be sufficient to minimise crossover distortion with genuine original output stage transistors but copies or fakes will probably need more, as the seller suggests by allowing for 100mA maximum. Older, plain epitaxial transistor types in copied cases could need a full 100 mA bias per pair, so that adds up to 100-200ma for each amplifier or 200-400 for the stereo pair. Good luck with the build(s).

BTW, a CFP output stage design can indeed use only 10-20mA bias per pair, but that's a different matter .
 
Last edited:
Hi,

the most important difference is, that Marantz use separated voltage amplifier, and power buffer stage to eliminate the back EMF distortion as much as possible.
So MA9 (and even SM11, and PM11) has two individually feedbacked stages. The output stage has a gain about 10-12dB only. The mentioned clone has the gain of 30.

I made the MA-9S2 kit and I like it quite a bit. For a fun project I was thinking of building another with the gain reduced to 10 dB and making up the gain with one of the clone Marantz JC-2 pre-amplifiers.

I think the gain is set with the ratio of the 2200 and 75 Ohm resistors. Perhaps I should use 2200 and 240 Ohms?

What do you think? Does this sound like a reasonable project and approach?

Perhaps it will get a little closer to the original? I can not do much about the kit not being balanced. But perhaps this proposal brings the gain and number of stages much closer to the original?
 
Last edited:
Not as easy. It may oscillate, because the compensation was planned for higher gain.
Some simulation may help to check, if it is possible.

Sajti


Yes, that is a real concern.

I was wondering how the MA-9S2 DIY compensation was done.

All I see as possible compensation components are the 10 Ohm and 220 pF which I have marked in the enclosed schematic and layout. Am I correct? Are those the components used for compensation in this design?

I wonder if anyone has a better schematic for this. I wonder if anyone has done any simulations on this design?
 

Attachments

  • How is MA-9S2 DIY Compensated Schematic.png
    How is MA-9S2 DIY Compensated Schematic.png
    337.8 KB · Views: 607
  • How is MA-9S2 DIY Compensated PCB.jpg
    How is MA-9S2 DIY Compensated PCB.jpg
    199.2 KB · Views: 562
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.