ExtremA, class-A strikes back?

Status
Not open for further replies.
Jump to Latest
#145
I'm on the same boat with you Christian, collecting parts. I've been busy with my job and not much time to do audio. I'm just collecting boards right now from different group buys and start working with them next year.

Thanks Jacco for some good insights, opinion and advices. I really appreciate it.
 
#147
Why not use the more linear Toshiba 2SC3281/2SA1302 instead, over the Sankens? You can parallel a pair of them.
Click to expand...

Linearity is not the primary concern here, this is a class A design with up to 75W of dissipation per output transistor. Hence these need to be able to handle the dissipation, have a large SOA as well as good characteristics. There's not many transistors that fit this application. The 2SA1216/2SC2922 have been extensively modelled and simulated prior to building up the output stage for verification. They provide a very good compromise in terms of maximum dissipation allowed, linearity, SOA and price/availability.

I'm sure there's better transistors to be found, however this amplifier was designed to provide the best possible performance by using standard parts. The math that governs this design determins performance for the most part, not the parts used.

Cheers,

Sander.
 
#148
projects update

How are all your extrema projects going?
My project is progressing very slowly, not much time due to work.
(Hoped to get more time off after my job change)🙁
Received today 90% of my components from Farnell, and hope to get started next month with stuffing the PCB's.

I want to order my 500/800VA toroidal transformer at:

http://www.schuro.de/

I know that it is possible to get the transformers made with a second voltage output.
Is it possible to get my 25v/30VA from a secundary voltage from the same transformer. Is this wise to do, or is it better to get 2 separate transformers.

Want to save some space in the chassis.
 
#151
See KSTR Post #77 for original data. KSTR's simulation distortion information shows very low total distortion, but higher percentage of odd harmonic than even, especially upper odd harmonics. The ExtremeA supersymm topology cancels out even harmonics, leaving a higher percentage odd harmonic content.

Has anyone run careful listening comparisons of this ExtremeA design against a single ended topology with low odd harmonic distortion? This ExtremeA amp would provide a good test of the well debated topic: Do amps with very low measured distortion sound better than amps with modest even harmonic but very low third harmonic plus very low upper odd harmonic distortion.

Does anyone have an LTC Spice schematic file for ExtremeA they can post?
 

Attachments

  • high_odd_harmonics.jpg
    high_odd_harmonics.jpg
    74.2 KB · Views: 1,334
#152
Does anyone have an LTC Spice schematic file for ExtremeA they can post?
Click to expand...

LTspice and Spice were used during the development phase of the ExtremA as the amplifier was extensively modelled prior to building up the first prototype. I'm not posting the files here for obvious reasons.

Having said that I do need to comment that using LTspice, or any other simulation, for a comparison between amps is a fool's errand. You can't compare amps based on a simulation that doesn't take into account all the issues you'll run into in practice.

It is simply another tool in the toolkit that allows you to achieve your goals more quickly, as less prototypes need to be constructed and optimized, but it is hardly conclusive. Besides the simulation's results are greatly influenced by the accuracy of the models used, I spent considerable time making sure that they were. Asking just 'someone' for a LTspice schematic file is a bit of a gamble, as who says the models used are accurate?

Cheers,

Sander.
 
#154
Boards Ordered

Large heat sinks are difficult to locate so I have been hoarding a lot of them over the years as I just discovered by going through the garage (after taking a few yrs off from DIY). So what to do with my heat sinks? Build this of course.

Boards are ordered.

If anyone has a digikey or mouser parts list please let me know via email or a post here. If not I'll create one.
 
#155
My parts list is almost complete.

What would happen if I used 16VCT transformers instead of the 18VCT transformers listed?

I suspect less 8 ohm power but would the SOA be better for 4 ohm use?

(only asking because I have 2 16VCT transformers that are 700VA.)
 
#156
Re: Boards Ordered

lgreen said:
Large heat sinks are difficult to locate so I have been hoarding a lot of them over the years as I just discovered by going through the garage (after taking a few yrs off from DIY). So what to do with my heat sinks? Build this of course.

Boards are ordered.

If anyone has a digikey or mouser parts list please let me know via email or a post here. If not I'll create one.
Click to expand...

Hi lgreen, you would surely help me by sharing your partlist. The building of my ExtremA is going very slowly.

Best regards,

Chris
 
#157
lgreen said:
My parts list is almost complete.

What would happen if I used 16VCT transformers instead of the 18VCT transformers listed?

I suspect less 8 ohm power but would the SOA be better for 4 ohm use?

(only asking because I have 2 16VCT transformers that are 700VA.)
Click to expand...

Looking at the schematics it seems that the bias currents are set with active sources, so even if you reduce the rail voltage nothing will change, other then the fact that output transistors will saturate sooner. So, yes, you will lose a bit of power, but that is it.

I am going in an opposite direction, as I have 4 units of 500 VA, 22 V centre-tap transformers here and I intend to use them. That would give me way more then 100 Watts (8 Ohm) for sure and it will make an amp a really HUGE.

I am thinking of biasing the amplifier for about 60-70 Watt Class A operation and once this level of power is reached, the amp will transit into Class A-B. I hope I am right about this. Opinions????
 
#158
Vadim,

I am thinking of biasing the amplifier for about 60-70 Watt Class A operation and once this level of power is reached, the amp will transit into Class A-B. I hope I am right about this.
Click to expand...

Yes, although distortion will also be significantly more, but still not in the same category as a typical class AB amplifier though.

Cheers,

Sander.
 
#159
SSassen said:
Vadim,



Yes, although distortion will also be significantly more, but still not in the same category as a typical class AB amplifier though.

Cheers,

Sander.
Click to expand...
Thanks Sander. Indeed the distortion will go up, but its ok, as the amp in my application, will rarely get to deliver more then 50-60 Watts anyway on any sustained basis. I am planning to use it with my RD-75 drivers from B&G company. THey are 75 inches tall line-source planer magnetics, rated at 6 ohm and are mostly resistive load. Have you heard of them?

I figure that the added A-B power (past 50-60 Watts) will provide the occasionally needed headroom.

Vadim
 
Status
Not open for further replies.
Top Bottom