Simple Symetrical Amplifier

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Did-you forget the inertia of the cone ?
And remember, if you send a high level 22µs pulse in-it, you will get awfull cone fractioning effects (distortion, etc...).
More than that, if you change the emitting surface at each level step, you will change the dynamic response curve a very strange way.
What about a plasma speaker, there is no inertia and and no frequency limit.

Sorry we are getting off topic.
 
Will the plasma speaker not produce the jaggies even more profoundly? I mean, plasma or whatever, the transducer has to move from zero to say 10 in a microsecond, then go back to zero again and then the cycle repeats.

My question is how, if possible, will the transducer interpolate the steps? Will the speaker throw a 44.1KHz square to us, with the sine hidden in it?

I have no knowledge of class-d, but these things just came to mind. :)

Anyways, I'm now in my friend's house arranging for a musical evening and will visit my village tomorrow!

How's cambe's and abetir's SSA going?
 
diyAudio Chief Moderator
Joined 2002
Paid Member
I had suggested that some time ago. And said that Class D, T etc. is the future.
Because it is near the digital domain, and, we can love-it or no, but digital will take all the audio market. We can imagine a direct oversampling/undersampling power converter, driving directly loudspeaker.
Because is is less Dependant on the qualities of the parts, so more industrial. So less expensive. and depend mainly on the design's intelligence and care.
Because active devices goes more and more fast; so it will be more and more accurate and easy to filtrate the switching frequency.
Because we are too many on this planet, and energy will become a big problem.

I have an idea, paralleling class D devices, with clock delays (oversampling) , on how to overpass the switching frequency limitation witch is the major problem of class D amps for the moment.

Have you listened to this French design? Talks some analogue steering of the class D. Devialet D-Premier, High End Audio Performances - DEVIALET
It was in the local audio show last week. 12K Euro I think.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
SSA going?

My friend's point to point one is going well he tells me. After he bypassed the 2X1N4148 analogous IR LED with Black Gate capacitors he had kept from older days, the sound opened up. He also had to drop the current in the BD transistors because they opened at about 25mA getting hot. He put 270 Ohm instead of 220 Ohm from base to rail and the spare 220 Ohm he put in series with the 1K2 current feed resistors. Now he got ~10mA on the drivers and they cooled down enough. He never gets more than 30mV offset after it had played for long or stayed on overnight. I hope I will visit him in next week to listen and to show you photos even. Have happy holiday!
 
dear shan,
I have etched this pee cee pee .but it is hard to me make holes .I tried to make holes but it is not finished .thanks for giving pee cee bee image(COPPER TRACK) thanking you yours ever masood
 

Attachments

  • Photo-0015.jpg
    Photo-0015.jpg
    895.2 KB · Views: 456
  • Photo-0016.jpg
    Photo-0016.jpg
    389 KB · Views: 428
Hi everyone,
the lateral I've order arrived today...
Sorry for repeating this question
which one should I make?
I'm thinking Nico's SSA & Shaan's SSA
what is the big difference of it?

Shaan maybe I will combine your SSA with Nico's one, please give your permission..?
307097d1350553331-simple-symetrical-amplifier-picibi.gif


If it is success, it can be used for two or more schematic in single board...
I need some input please :)
I will make my own board, I will try as symmetric as I can do.

Regards
 

Attachments

  • J162 K1058.jpg
    J162 K1058.jpg
    164.4 KB · Views: 396
Last edited:
Hi John,

I think Shaan's amplifier has gone through much more refinement than mine did a year ago and was essentially used to evaluate quickly what the Cat dragged in.. There are a few problems I have noted which is can be summarised as follows:

The driver transistors should best be BD139/140, the small signal device2SA970/2SC2240 just smokes at higher voltages but performs fine at 24VDC.

I have also noted problems with the 5K pot forgetting that this amp is symmetrical and the adjustment of off-set is very small. in fact only tens of ohms off the 1K mark, thus starting at a pot setting of 0 Ohms has catastrophic results. :eek: and possibly pop every device on the PCB

The pot should be set to 1 K and small adjustments made left and right for those off-set critical maniacs. In fact I don't know it offset adjustment is really that necessary and I used a fixed 1 K resistor which worked fine for me.

I have changed the 330 Ohm resistance between MOSFET gates with a 1N 4007 (any rectifier diode will do) and it tends to track any ambient changes well keeping offset more stable than with the resistor.

The bias current with the diode is around 180 - 210 mA and optimum for these output devices, but one must keep in mind that this amplifier completes a full circle so to speak, adjusting anything has an equal and opposite effect elsewhere. For instance changing the gain setting shunt resistors impacts on the output bias current. Everything is intewoven and you must keep your mind clear of distractions

My amp is still the crows nest it was made a year ago and I believe that much refinements has taken place since the very early stages, thus I would recommend Shaan's refinements if he is comfortable with my vote of confidence. Mine needs some experimenting from the builders point of view and is not absolutely optimized but will gladly assist anyone who wants to tackle it from his own perspective or how he wants to tailor it.SSA is in effect a very simple symmetrical design but the emphasis is on symetrical not on the fact that it looks all balanced as drawn, The whole amplifier has to be in equalibrium or nothing, there is no half good.

To summarize and I hope Andrej does not object - SSA is both simplex as well as complex, it is for the DIY that wants to experiment and get things the way he would like to have them.

My SSA as driven nothing other than headphones and on occasion BBC LS3/5a monitors very successfully, but it was tweaked for my purpose and being supplied by 24VDC rails.
 
Last edited:
Hi John,

The driver transistors should best be BD139/140, the small signal device2SA970/2SC2240 just smokes at higher voltages but performs fine at 24VDC.

I have also noted problems with the 5K pot forgetting that this amp is symmetrical and the adjustment of off-set is very small. in fact only tens of ohms off the 1K mark, thus starting at a pot setting of 0 Ohms has catastrophic results. :eek: and possibly pop every device on the PCB

The pot should be set to 1 K and small adjustments made left and right for those off-set critical maniacs. In fact I don't know it offset adjustment is really that necessary and I used a fixed 1 K resistor which worked fine for me.

I have changed the 330 Ohm resistance between MOSFET gates with a 1N 4007 (any rectifier diode will do) and it tends to track any ambient changes well keeping offset more stable than with the resistor.

The bias current with the diode is around 180 - 210 mA and optimum for these output devices, but one must keep in mind that this amplifier completes a full circle so to speak, adjusting anything has an equal and opposite effect elsewhere. For instance changing the gain setting shunt resistors impacts on the output bias current. Everything is intewoven and you must keep your mind clear of distractions
Thanks Nico, great point there...
But it is quite mysterious why 2SA970/2SC2240 fail at higher voltage, maybe too high.

I will at least supply +/- 35VDC max to +/-50 if possible??? Need advice here...
Sometimes ago I have modified BJT amps that run +/-35VDC to +/-45VDC but why the lower voltage give better sound, I hope to this time I build SSA give better sound with a little higher voltage :)

So for some adjustment it just like we adjust 2 amps at same time, Ok I will very careful :D

The Laterals shown in the picture is a bit suspect. They could be fake or at best Chinese; the latter cannot handle the same current or dissipation as the Hitachi originals.
Hi Samuel, I'm a little suspect too...
but I believe they are original one, I trust the seller but don't know where it is came from...:D
btw they look just like ordinary transistor not special looking :(
I more like TO-264 & the big MT-200 :cool:
just let it sound decide :note:


Regards
 
Hi Samuel, I'm a little suspect too...
but I believe they are original one, I trust the seller but don't know where it is came from...:D
btw they look just like ordinary transistor not special looking :(
I more like TO-264 & the big MT-200 :cool:
just let it sound decide :note:


Regards

hi john
i use this mosfet for sigma PA by dr. bora & minicrescendo PA, n everything is ok...
mr. naf also use this mosfet for his SSA project ....
this mosfet not from hitachi but from renesas
cmiiw
 
Hi John, it is a question of power dissipation on the driver transistors, unless you are prepared to make other modifications such as adding emitter resistors of about 150 ohms and a few other minor adjustments they can be made to work in the application.

Only thing is you must not be a novice else you will run into problems. I used 2SCX2240/2SA970 rather than the medium power devices purely because I could increase slew rate and decrease simulated THD - no other reason.
 
hi john
i use this mosfet for sigma PA by dr. bora & minicrescendo PA, n everything is ok...
mr. naf also use this mosfet for his SSA project ....
this mosfet not from hitachi but from renesas
cmiiw
Hi Danzrose, glad to see you here...:)
No I'm not doubt about it genuine
& they all arrived save, only one leg bent :D
they just a little bit small that all

Do you want to build SSA too, right?
Which one do you wanna go?

btw Thank you Danzrose

Right now I'm reading whole thread
looking something I must know
that all almost 3500 posts :eek:
 
Last edited:
My friend's point to point one is going well he tells me. After he bypassed the 2X1N4148 analogous IR LED with Black Gate capacitors he had kept from older days, the sound opened up. He also had to drop the current in the BD transistors because they opened at about 25mA getting hot. He put 270 Ohm instead of 220 Ohm from base to rail and the spare 220 Ohm he put in series with the 1K2 current feed resistors. Now he got ~10mA on the drivers and they cooled down enough. He never gets more than 30mV offset after it had played for long or stayed on overnight. I hope I will visit him in next week to listen and to show you photos even. Have happy holiday!
thanks for the info sir, i'll try it too
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.