sp300b

sdravoNice work Sp Do you know the gain of SA1837? Just to have an idea of amplification ratio...(Sorry if this subject was discussed in Serbian ... me just understand zeleznicka stanica )It would be nice to test the TL431 bias vs Cheffs simple but efficient vbe multiplier. Do you intend to do that? Looking forward...Manu

Zdravo Manu.I don't have an idea about amplification factor of 2SA1837.But it's a part of current mirror, and at the end we will fix the amplification of the whole UGS by means of feddback/input Resistor ratio ;-)TL431 as well as the whole circuit will be made and serious testing will be done. Stay tuned ;-)ps. What happened to Your Pumpkin - finished ?

sad sam ti poslao mail sa korekcijama ;naravno , opet ne gledaj vrednosti naravno - zaboravio pominjani cap od 4u7 ;take it easy - ako imas gde - turaj ga - i racunaj footprint za 4u7/25V elko - ne treba nista fancytelefonom mozemo dogovoriti sta od otpora pobacati sa UGS pcb napolje (na MB), a da zadrzimo najveci stepen versatilitytreba uzeti u obzir da UGS pcb ionako nije autonomna ...

Vazi sutra/prekosutra...laku noc svima

yup

Danke...

Work in progress...Vrednosti elemenata tek treba izracunati.Any problem so far?Serbian_XA_sch_rev_D.pdf

Pogledaj ovde - mozda nadjes nesto

cisto kao podsetnik - izvadak iz leaving class a :

Podsetnik :

Napajanje = plus/minus 18 V

Mirna struja 1.5 A po grani (ukupno 3)

Po jedan P i N mosfet po grani

Bez Source Otpornika

MJL naruci iz digikey-a jer je jeftiniji a sto se tice samostalnog narucivanja, cesto sam razmisljao ali kada obracunam postanske troskove i spediciju i carinu pa jos izgubljeno vreme, jeftinije je da mi neko od dobavljaca donese i ja samo pokupim.

Ja se u potpunosti slazem. Elegantno se narucuje (putem Adobe formulara i placanje VISA karticom) recimo sa Partsconnection, ali carinu neces moci da izbegnes. Ja sam u dva tri navrata uzimao robe do 100 Eur. Isporuka ti stize za 7-10 dana.

http://passlabs.com/awards.htm

Pozdrav svima

Pogledam na audioholics-u uopsten komentar i preporucene dilerske cene pojedinih modela od XA30.5 pa do XA200.5 i ...

pogledajte i samo ako ve

Nope... to ste ti i sp300b(desni)!

Bolje ti je da zavrsavas skolu (i babela i LDR)...

I'll take it...samo gledam Passove hladnjake na XA30.5 i cudim se kako mu izdrzavaju...Mora da se greju na nekih 75 stepeni, a on tvrdi da je 53

p.s. edit: citat iz passovog clanka "Leaving..."

As a practical matter, this means that our X (Class AB) amplifi ers are biased to

dissipate roughly half of their rated output power. The XA (Class A) amplifiers are

biased to dissipate roughly three times their rated output power.

Da li ovo znaci 3 x 30 = 90 W?, a ja se "patim" i pokusavam da se uglavim na 100 W.... ili da ja malo odmorim mozak kojih dan-dva (ili nedelju/dve)

A klasu odredjuje struja kroz izlazne tranzistore. Za 30W u a klasi na 8 oma struja bi trebala biti koren i 3,75 a to mu dodje 1,94A. U varijantu koju Vi zelite od 25W na 8 oma struja bi trebala biti koren iz 3,15 a to je 1,78A. malim smanjenjem napona napajanja i 10% manjom strujom mozda bi smanjili disipaciju na 100W. Sa 1,78A bi napon anapajanja trebao biti +/-14V da bi disipacija na jednom pojacavacu bila 50W odnosno 100W na bridge kanalu, sto je jako mali napon da bi fetovi radili ispravno.

Sve se slazem. P = R * I^2.

Sto se Disipacije tice, ona je jednaka naponu napajanja (u odnosu na gnd x 2) i jos puta 2 struje (levo i desno).

Ako pricamo o napajanju od +/-20V onda imamo zahtevanu struju od 1.25A po "grani" bridge pojacavaca za 100W disipacije

Korisna snaga u tom slucaju iznosi R* I^2 = 12,5 W...a ja razmisljam o 25W ...

U svakom slucaju, hvala VAM kolega Impulse na zelji da pomognete.

Pretpostavljam da prica o snazi u A klasi dok je u AB klasi snaga mnogo veca

I NP i ja pricamo o push-pull u A klasi, samo on spominje 30W, a ja 25 W...uz max dispaciju u mom slucaju od 100 W na hladnjaku...

25W u bridzu bi zahtevalo jako mali napon pri kom fetovi ne bi hteli da rade.

Ok - kontam, ali kako onda Pass deklarise XA30.5 kao 30 W?

Potrebna mi je konsultacija.

Naime, razmisljajuci o disipaciji za XA.5 Concept, a uvazavajuci NP objasnjenje, te obzirom na tabelu datu u dokumentu Leaving class A, pokusavam da shvatim da li mogu na racun manje korisne snage (nesto tipa 25 W) da idem sa disipacijom od 100W (pricamo za jedan Mono kanal)?

Naime u svom Babelu imam hladnjake koji se regularno temperaturno ponasaju pri cemu trose 100W disipacije.

- Racunajuci na identicne hladnjake, koji bi to trebalo da budu moguci Naponi i biasi ?

Na kraju krajeva, ja sam nesto skontao da Pass u XA30.5 koristi plus minus 23 V, ali ne razumem bas sa kojim biasom?

Details

Cascode

The cascode therefore, are BJT transistors inserted in series in the drains of the JFET. Their bases are maintained at a constant Voltage potential. As the potential of the base is constant (in the order of 10V for UGS), potential of the Drains will also be fixed, and 0,7V will be less than the potential of the bases. What makes that is ultimately drains the JFET will be at a constant potential. The VGS of JFET in normal operation is relatively low, we will quickly say that the VDS voltage at the terminals of JFET remained stable, or at least varies very little.

With this little trick we have three very interesting conclusions:

First, that the voltage of the JFET is within the 10V, that means we have reduced the power dissipated in the JFET: for example, for a current of 3.5mA, a JFET will dissipate 35mW, while without cascode, it should remove more than 65mW. It has therefore gained in overall thermal stability.

Second, as VDS is roughly constant, it will have the effect of linearization of the behavior of the JFET.

For an ideal transistor, the various "rights" to VGS = CSTE are parallel and slope zero. In reality, they are not, and the slope is not zero. So if the voltage of the drain JFET varies, the current drain of the JFET will also vary ... And where is the problem, you say? Well, is that we would like the current drain of JFET depends only on the input voltage, and not something else. (I remind you that had asked id = gm*vgs)

The cascode, setting the potential drain, thus to minimize this non-linearity, and working on a much closer these curves (in red zone ☺), where you can take that will variation of drain current negligible according to VDS.

Okay. Purists will tell me that VDS is to be maintained constant, and not only the drain voltage, as is done in the hope that the voltage source does not move too much. That's true. But I have not wanted to break the head, and you are free to try both versions, after all.

With all that, I forget to mention the last effect of the cascode, the extension of the bandwidth. B

itd...

Da se ne zaboravi ...

Btw - odlicno objasnjenje za siroke narodne mase koje je sheff uradio.

nazalost na francuskom samo...za svoju dusu sam pokupio i google prevod na engleski, pa pokusavam da uklopim celu pricu...

Ma slikao sam ja uspravno ali sam aparah drzao horizontalno

Topla preporuka za sve moguce marufetluke sa slikama - FSViewer i ne boli glava

Zato su tu tzv helperi

Nedostaje sta ?!

Dobrodosao u klub Babelovaca

Nameracio sam se da izmerim Vbe za potrebe uparivanja tranzistora, trazio informaciju na Internetu, a posto nesto bas i nisam naleteo na istu, seo i nacrtao slicice koje objasnjavaju.Stoga, evo neka ostane ovde zapisano.(U mom slucaju, hteo sam da umerim na Ic = 5mA).