NMOS for final stage UcD

On a sunny day (Wed, 7 May 2014 08:41:41 -0700 (PDT)) it happened Vlad wrote in :

Searching on ebay for 'power MOSFET nC' eBay item number: 171084598059 Power Mosfet 8N80 8A 800V 35nC 13pF DMOS insulated FQPF8N80C 2$90 cents probably does not meet your current requirement?

eBay item number: 181187201164 Power Mosfet P-CHANNEL FET FQP3P50 500V 3A 18nC 9.5pF Fairchild USA Qty:4 5$90 :-) Use 16 Ohm speaker?

Well, I did have an idea of making an audio-modulated high frequency welding machine... you know, combining work with pleasure :-)

Den onsdag den 7. maj 2014 15.04.09 UTC+2 skrev Vlad:

an other > 400 kHz oscillator in the room.

At full modulation it drops to as low as 86kHz, zero crossings are around 2

80kHz. Even so the THD with plotwinsize=0 and numdgt=16 says 0.0924%@21 W/4R (efficiency is 93.37%). It's also true the sources are modelled with R ser=0.1 Cpar=1m and parasitics near the end stage (on all pins, D, S, G ) are 25nH Rser=25m. Output filter is a Chan core inductor and the capaci tor has Lser=25nH and Rser=25m. Could these mean I can go for some chea per (higher) Rds FETs? Last choice is a PSMN017-30BL and the dissipated pow er is ~250mW. B I

I seem to remember IRF pushing one of their directfet for class-d audio IRF6665 I think

-Lasse

The specs look impressive, the price is a bit high but this may not be the choice because I really don't know how am I going to solder that. Checking IR at their class-D shows a TO220 cousin, IRFB5615, which seems more mallea ble, even if it's over-rated spec-wise. I'll keep looking, though.

Or Up Counter Down (that's infix notation).

Ok, when I went to school, Q9,10 were NPN not PNP (flip the E + C ) Maybe already spotted. Cheers,

Harry

I remembered I have seen somewhere in their pdfs a multiple feedback input, as opposed to the simple pole-zero, so I tried it, too. I got the frequenc y between 157kHz (max modulation) and 340kHz (zero crossings). The efficien cy is also lower, 91.87%, and the distortions are larger, 0.23%@18W/4R. FET s also dissipate 0.41W each. I think I liked it better before.

Digging on the net I found their newest technology, the ncore:

. Looks like a good read.

What is the distortion at low levels, say 100mW? ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

I can't speak for their products since I have never had/heard one but, if you want, here's their products page --

ct-downloads.html -- and the first link, NC400, is the kit made for diy-ers . The datasheet has some graphs there that made me go mute for a while. See for yourself.

UcD?

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com 

Precision electronic instrumentation 



You guys are all blind, Q9 + Q10 should be NPN. What is going on over  
there?? 

Cheers,  Harry

I think it's meant to be so, it's not a totem-pole, it's current-driven by Q7/Q8. The way I see it, when Q8 conducts, D9 conducts which blocks Q10 (ba se goes low together with collector) and puts Q12's base to Q8's level thro ugh R20, which discharges Q14's gate. When Q8 is closed, Q12 is blocked by D9 (nothing to discharge anymore), Q10 opens and discharges the bootstrap C

5/C6. I could be wrong, of course.

[snip>

Brilliant! Good for Alu TIG welding :)

On a sunny day (09 May 2014 14:18:19 GMT) it happened joe hey wrote in :

Maybe, but I think the idea of audio PWM for home power applications makes little sense. the AVERAGE power in music is really low, and a good linear class B will do. the average power in speech is about 20%. So he should not worry about dissipation of a few percent of 400 mW in a DPAK. The only time PWM audio (class D) makes sense is in ultra low power applications like cellphones to extend battery life (and that gives little interference), or in extremely high power apps like big halls public address perhaps, provided every thing is screened. That design leaks like a sinking boat, the few hundred nF caps at the output have RF on them, are way to small for the lowest frequencies (< 100kHz) to do anything, it is frequency modulated by the audio on top of that (so makes sure it messes up EVERY station on MW and LW and SW (harmonics) ) [1], and for 10$ you can buy a good PWM module on ebay that sucks way less. And speakers have 20% distortion all by themselves so why the hype Just use a good class B normal amp for music and be done with it. For welding? Well sure try it, he is far away from here, should not interfere much. My as usual not humble opinion, but technical views may hurt the .philes but stick it up ..

[1] do not start on the spread spectrum jive.

Portable TIG welders already are PWM equipped. And have HF 'starting' bursts. With respect to FM I think you're right and PWM would do better.

joe

e sense.

do.

That would make sense for speeches, not music, wouldn't it?

DPAK.

I'm sorry, but that's not the case here. The transistors keep on switching even for zero signal, unless the mute function is used (or some signal dete ction thingy), so those few hundred mW are real even for 1% power (though n ot quite so at that level).

cations

e),

Agreed, efficiency being the trump card, but this UcD/ncore technology seem s a bit different since they claim, with their ncore, that the best class A B can't match it.

or the lowest frequencies (< 100kHz) to do anything,

esses up EVERY station on MW and LW and SW (harmonics) ) [1],

Well, that app note is made as a showcase for the general public, I don't e xpect it to give ultra HD sound or even >90% efficiency in real life. My go al is to reach those figures based on that schematic.

Yes, but, it's one thing to have 20% x low distortion and another to have 2

0% x high distortion. The same goes for efficiency, it's enough that the lo ad itself is highly reactive and I don't get to see the maximum RMS power t he amp can provide, so I may as well compensate however I can. Besides, for power amps (> hundreds W), it matters (not my case, too many neighbours).

fere much.

I'm not an audiophile and I certainly don't want to have a fight (verbal or not) as my goal -- and I hope you don't see it as such. I do appreciate th at people are answering since I know my level of knowledge is not the same. If I started this thread it was because I am interested to build this thin g and because the graphs on their products are extremely appealing:

ml

On a sunny day (Fri, 9 May 2014 08:19:46 -0700 (PDT)) it happened Vlad wrote in :

I do not remember the exact % for music, but it is really low too. Nothing like a 100% sinewave. IF you allow for some headroom. Given that some music is companded and always close to 80% (AM stations do that), it can be the other way.

Yes, right, but that would mean you have 1W (for example) at no signal! There goes efficiency!

Well, I have build many amps,. AB, cannot even remember the number, much more than 20 for sure, and distortion was never an issue, only issue being how much 'A' say quiescent current. Distortion is a very subjective thing, once I was so enthusiastic about the sound of my speakers until I moved house, and it sounded 100% different, room acoustics... Makes you humble. The piezo speaker in my boxes have audible distortion, I was looking for a quad electrostat, but cannot convince myself to fork out that much (for a second hand one), and I do no hear that high anyways anymore.

Aha.

But distortions do not multiply?

Yes same problem here, if I open my whatsit 300W stage amplifier it is like a moker hammer hitting the floor. Noticeable across the road where the floors and windows start swinging. So if I really want to listen to some pure good sound I use Sennheiser earplugs.. Now there starts an interesting thing about audio, these were way to loud on my shortwave receiver, it puts out several volts pp, I did a test for what my ears still felt as 'acceptable volume', and found it needed 30 x attenuation (as in voltage), also needed AC decoupling to reduce the power up thump sound... Anyways build as small adaptor, and uses 100uF and 1k in series with the 39 Ohm Sennheisers. Now that gave me the old tube sound! Kevin would like it! The 'tube sound' is basically high impedance drive of speakers, the 1k creates this, the damping of the earphones is less, the frequency characteristic changed, I expected it to be shit, but it turns out to be very pleasant for radio, and acceptable for music, So bit of audio alchemy.... Now you ..philes, 1 k in series with 100u in series with your 39 Ohm Sennheiser earbuds gives you the original warm tube sound, go patent it!

Well as J.L. noticed, datasheets can be made convincing in a way. I know you are serious about that thing, hope my posting did not offend you, or anybody, but a bit reality check as far as audio is concerned every now and then is a good thing in my, again, not so humble opinion.

ou, or anybody,

s a good thing in my, again,

As long as creative messages come from it, no offense for me, and my hope i s that the price to pay isn't your average argue. At the very least, making this darn thing will give me a practical grasp on it which can only go two ways, to put it simple: love it or hate it.

There isn't one Tchakoovski's "1812 Overature" has cannon in the pecussion section.

bagpipe, and organ music on the other hand can be closer to a constant power level.

--
umop apisdn 


--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

On a sunny day (10 May 2014 07:09:46 GMT) it happened Jasen Betts wrote in :

Yes headroom, that means everything else needs to be lower most of the time, the alternative is let it clip on the cannon, not good, so then the strategy is companding if you have low dynamic range, like for MW or SW radio transmissions: sox example.wav example_companded.wav compand 0.003,.013.5 -80,-40,-60,-30,-40,-20,0,0 1 imitates a NE570 compander chip, see my site:

So its easy by changing some parameters to get close to 100% all the time, but that is NOT HiFi.

Yes, sure, put some clicksons on the heatsinks in series with the power, That will teach them ;-)

While I understand that some people don't care too much about how things so und or that they are content with whatever results they have in front of th em, that doesn't mean other people have to do so, either. I'm also saying t his with the knowing that, in the end, the final sensor dictating the quali ty of the listening is one's own perception, which is a very moody beast an d particular to each one. While some people are alright with anything that doesn't sound too distorted, others will take care that the sound fits thei r needs. Does that make them audiophiles? Certainly not, that's just a fanc y term. But what does that make them? Nothing more, nothing less than human s, just like anyone else, but with the difference that they put more accent in how the sound comes out of their audio "conglomerate". Alright, fine, but then why the fuss about efficiency? Well, in some places , "waste not" is not taken too lightly (by me, at least), so if two results sound the same but at different power consumption, then there's no reason why one shouldn't follow that goal, is there?

I'm not an audiophile, I don't consider myself as such and I actually resen t being labeled so (not implying I have been, here), but I do have these tw o ears that seem to interpret sounds differently than others. Better? Worse ? Who can tell? It's just different and they are, in the end, the ones dict ating the outcome of whatever should turn out of this would-be amplifier.

Numbers, statistics, these mean nothing to the ear, not even to one used wi th numbers and statistics. The ears are the final stage in any amplifier an d it's them that will have the last word. Curse them!