Power amp under development

7 pairs, oh my god...

What are we doing with these amps, running them as
industrial controllers.?

SOA under normal music conditions would be plenty reliable
as Quasi mentioned earlier in his thread.

I run my 2 pair version (60-0-60DC) all day into a 4R0 load
without breaking a sweat.
This is DIY, let him build it, burn it up, and learn his
lesson. :devilr:

If he does want to try, he'll need this to burn it up with..

[url]http://www.antekinc.com/AN-15450.pdf
[/URL]

and use a water heater calrod
An externally hosted image should be here but it was not working when we last tested it.

to boil water and test SOA... 😀 😀

p.s. I will model a design ,but nothing beats a real world test .
 
Well well !!

Hi,

Andrew same story as mine !! but the reason for me to go for that 4 volt extra was to compensate the poor voltage condition here !! its always not exact 230V most of the times its around 220-225,Hence thought to add the extra buffer !

But however I did check the supply daily from the day you mentioned it in the BOQ thread,and to my utter surprise😱 the power is actually almost constant at 230V over it at certain occasions (did touch 240) !!

Now Am also left with 2 trans of 650VA (65-0-65) ! But I though wanted to use it for Bi-Polar version,but the Quasi rules is same for all I guess !!

I was thinking of getting the secondary reduced to may be 50V,but the guy says its like almost making new trans except for the existing core and primary winding !😡

So should I put it on sale here in the trading section ?? will there be any buyers for it ??

Also I wanted to ask one more thing ! I also plan to build the Nmos200 for the multi channel purpose ! And I have 2 set of speakers with me one 8r and one 4r (pair that is) !

So whats the ideal VA rating for that ? want it to be minimum of 150W @ 8 don't really care bout the 4 but should not die if connected and run for couple of hours !!

I know you've said it a million times that output power X 1.5= VA but was bit confused as I want to use it for both 8 &4 Ohm load (!!The voltage will be 35VAC (as it will be 49VDC) and will be well within the stipulated max 50V !!🙂


Please do throw some light on this !!😉

Regards.
 
I have yet to buy the parts (10/1/08!!) for my Nmos 200,s
but I have (2,for stereo) 400VA Fisher receiver transformers (45-0-45) which I believe would get me at least 150+ W per CH 8R0.
I have these right in front of me and get 61.5-0-61.5 DC.

Soundsgreat, sell yours and get:

http://www.antekinc.com/AN-6445.pdf 600VA

only 60$ US..good deal.They also have a 48-0-48v
(extra headroom):devilr: You can always upgrade
a transformer as long as you have good amps/caps/bridges.
(maybe change R6,no bc546 at inputs!)
OS
 
canca87 said:
smoke 😕

well the amp worked great at low volumes, worked for about an hour. As soon as i cranked it up, smoke. it seems that T9 did something wierd and applied source voltage to the gate's on the positive side. all positive fet's are now dead, T9 is so burnt there arent even pins left on it, and all the 27ohm gate resistors are cactus.... it seems the negative side of the amp is all fine thou...

im still stumped.. could it have been fake fet's? why did T9 go all wierd then? T9 bjt is from a trusted source, highly doubt it was a fake or dodgey. im at a loss... any idea's as to what could have caused that?

anyway, once i know what went wrong i will rebuild it. it was good untill i cranked it up. can supply pics and measurements if anyone wants.

Hi Canca87,

You should post pictures of your build to help us determine the failure. I note you weren't entirely happy about the thermal coupling of the output stage and this could be the cause. It only takes the failure of one FET to cause a catastrophic chain reaction.

Your description of T9 sounds like it failed after the FETs failed. A shorted gate source junction in one of the FETs would have forced T9 to drive the load on its own (bang!). It is also possible that the negative rail FETs are damaged so you will need to check these. This is possible because if the +ve rail FETs are shorted from drain to source then the negative FETs will try to conduct from +ve rail to -ve rail.

When you re-build, check that the bias current is stable. When the amp is running check that the heatsink bracket you are using doesn't get too hot. You should be able to touch it firmly without discomfort. The Nmos350 & 500 was designed to have the FETs bolted directly to a large heatsink. The Nmos200 uses a bracket but is limited to 50 volt rails.

Finally what rating fuses are you using?

Cheers
Q
 
ostripper said:
Quasi, I would like to know a little bit more about the
"birth" of the quasi line of amps. What sparked this interest
was the reading of "8 distortions" by Douglas self.As I read
I saw he progressed almost to your exact design topology
as the truely "blameless" amp.(Which by the way has
.008-.01 distortion and bandwidth out to 200k+)

My question is, did you read Douglas self's book to base
your design on and if so would it be a good reference to
further it's development. ?

thank you for a GREAT design 😀 ,OS.


P.S check out my "WWW" many books,designs, etc.
quasi-DXamp-leach are abut the same for total downloads...
(remember the old forrest mims-radio shack 1978 .. books for
curious children ? ..they are here in
electronics/educational/forrest mims/) 😎


Hi Ostripper,

I have not read any of Self's books altough I may have read articles by him in the odd magazine, I just can't remember. The design topology I have chosen is very common and variations of this is found almost everywhere. So the main IP in this amp I guess is the layout and the choice of an all N-channel FET output stage.

The main design objective for the Nmos series was to put commonly available components to good use and as you have found, keep the price down. Most people will spend more money on the power supply than they do on the amp modules.

The Nmos series is not by any stretch of imagination one of the best power amps in the world, but more a practical demonstration of what can be achieved in a collaborative forum. You'll note that the final design is quite different to the original proposed, and when I get around to it the Nmos MkII will be a further refinement.

Cheers
Q
 
ravslanka said:
hi quasi

now iam going to make nmos 500
is it ok to drive it with 65-0-65 vac 7amp in 4r load

regards


Hi Ravslanka

If your transformer is a torroid then it migh be simpple to unwind some turns off the secondary. The transformer should be wound bi-filar (both wires wound together) so find these and take about 10 turns off. This will reduce the AC voltage to a suitable level for a Nmos500.

The other thing to consider is whether the transformer voltage of 65 0 65 vac is measured unload or loaded. Calculating your DC rails should be done with the unloaded voltage of the transormer otherwise you will get higher than expected DC rails.

If you want to build using the 65-0-65v transformer then consider the Actrk600. This amp is more efficient (less heat) and has room for 6 FETs per rail. Remember though you will have DC rails of between 90 to 95 volts (loaded / unloaded).

By the way I do have an early version of the Actrk600 running 93 volt rails myself.

Cheers
Q
 
Yeah , boiling water with a BOQ*, it is shocking to some
people but sells them on the DIY factor.:devilr:

A 30Vrms sine makes the water hot,element resistance goes down
(spot of tea ,anyone.?) Sure saves on voice coils.😀

I will post pictures on 10/6 or later "qausi showdown,
BOQ* vs. NMOS" we'll see the which one makes the tea
faster...😀 😀

BOQ*= brother of quasi.
 
Hi,

Andrew and Ostripper,Thanks for the advice !

But Andrew I know the VA for the 8R and have almost selected 350VA for the same ! the only thing I wanted confirmation is this

So whats the ideal VA rating for that ? want it to be minimum of 150W @ 8 don't really care bout the 4 but should not die if connected and run for couple of hours !!

So wanted the confirmation that with the 350VA or 300 as you've suggested,the amp will drive a 4R load comfortably ?? will there be a loss in sound quality ?? and other common things associated ??

The most you can expect from a BJT amp is 110W and from a FET amp around 90W.

Now what do I do or how do I implement a BJT in Nmos200 ?? ok if I forget 150,atleast say 100 +W with the existing devices will I be able to get ?? say increase the AC to 40V (56V DC) or have 45 as ostripper says and get 150W ??

I may sound stupid asking all this but I don't want to repeat the same mistake and end-up losing 2 more trans😱 !

Again I'll repeat,I will nominally use 8R but occasionally may use 4R for couple of hours (say to watch a movie or listen to couple of CDS)

So keeping both in mind ! is there middle way ?? or any other alternatives ?? I cannot spend on a 600 or more VA trans as the cost will kill me !

Also as I said earlier Am making a BI-Polor version(for which Am ok with going for a 600+VA Trans) so this 200 is for a second set-up or alternative set-up or anything that one wants to call it !!

Regards.
 
Now what do I do or how do I implement a BJT in Nmos200 ??

It is easy to do just replace the "front end" (MJE 340/50 ,outputs)
of the NMOS 200 with the "front end" of the BJT "brother of
quasi' (MJE15030/31 , 4-MJ15003 outputs), and you are
good to go . For 50VDC use NMOS 200 bc546 trannies/R6=15K,but for 60-70V use 2sc1845 for LPT and
R6=18K (more like NMOS 350).With your 40-0-40 you
will get 120w at 8R0 ,at 4R0 you might load down your
supply.(best to use 2sc1845, it makes amp more universal,
any supply can be used 40-55VAC,just change r6 using
Qausi's chart.)

P.S. I use it to drive a Dayton 4R0 12"sub. 🙂
 
quasi said:



Hi Ravslanka

If your transformer is a torroid then it migh be simpple to unwind some turns off the secondary. The transformer should be wound bi-filar (both wires wound together) so find these and take about 10 turns off. This will reduce the AC voltage to a suitable level for a Nmos500.

The other thing to consider is whether the transformer voltage of 65 0 65 vac is measured unload or loaded. Calculating your DC rails should be done with the unloaded voltage of the transormer otherwise you will get higher than expected DC rails.

If you want to build using the 65-0-65v transformer then consider the Actrk600. This amp is more efficient (less heat) and has room for 6 FETs per rail. Remember though you will have DC rails of between 90 to 95 volts (loaded / unloaded).

By the way I do have an early version of the Actrk600 running 93 volt rails myself.

Cheers
Q

hi quasi

thanks for your advice

problem solved

i made a new transformer with 60-0-66 9 ampere ei core
now i will continue with my amp
iam using an ei core transformer
 
Hi,

Ostripper Thanks a ton for the advice :worship: ! almost gives me all the answers I was looking for !

Again just a small doubt ! please bear with me !

at 4R0 you might load down your supply.

What exactly you mean ?? I have some idea but want your version ! please excuse if I sound noob ! but with this Amp am going crazy day by day (nothing because of the design or the likes of it) !!:headbash:

anyway as I said earlier in my post Am yet to order the transformer for the Nmos200 ! so have to do it based on your recommendation ! I might chose 45VAC (63VDC) ! try to achieve 150W like in your earlier post !! The VA rating will remain 350 !!

Regards.