Tea-Bag Napisano Jun 3, 2010 #243 Napisano Jun 3, 2010 :ok:Zen Mod,Did you use matched R100's per channel?
Zen Mod Napisano Jun 3, 2010 Autor #244 Napisano Jun 3, 2010 :ok:Zen Mod,Did you use matched R100's per channel?not exactly matched you can see their Vgs-es at 2A on pictures
Tea-Bag Napisano Jun 3, 2010 #245 Napisano Jun 3, 2010 : Thanks, do you think matching is needed or best practice here?
Zen Mod Napisano Jun 3, 2010 Autor #246 Napisano Jun 3, 2010 just best practice but - I didn't tried really mismatched ones ; maybe I will , but not so soon ...... maybe because I don't have really mismatched ones I know that they're everywhere from 1V2 to 1V9 , at 2A in any case - I think that real and valid answer you can get only using THD meter , and comparing best and worst case
Zen Mod Napisano Jun 3, 2010 Autor #247 Napisano Jun 3, 2010 Tea- just few thoughts ...... as you can always see from Papa's schematics - he always counts on selected parts ; same is here ; value of drain resistor in input LTP is dependable of exact lower output Jfet Ugs value ; same - value of resistance in paralell to CNY bjt is dependable of exact upper output Jfet Ugs value . so - procedure for making functional Babelfish J2 is something as : 1. test and choose output Jfets ; according to their respective Ugs values - 2. preset input LTP resistance and CNY's parallel resistance 3.preset input LTP CCS at 20mA ; then 4. solder outputs in place and proceed with proper power up procedure #3 demands some tricks how to , with auxiliary PSU ...... or just setting programming resistance in CCS to ~ 0V65/20mA=32R5 more to come - with proper schematic , with parts nomenclature
Vix Napisano Jun 3, 2010 #248 Napisano Jun 3, 2010 lako vam je mene zekiti ........ Pa da, kad se ima za cega u'vatiti Ma, glavno da igracka radi...samo da se ne pocne pusiti...
Zen Mod Napisano Jun 3, 2010 Autor #249 Napisano Jun 3, 2010 Pa da, kad se ima za cega u'vatiti Ma, glavno da igracka radi...samo da se ne pocne pusiti... nece - ne boj se - ene je rastavljena ce sledece pravimo Little Donut
Tea-Bag Napisano Jun 3, 2010 #250 Napisano Jun 3, 2010 Tea- just few thoughts ...... as you can always see from Papa's schematics - he always counts on selected parts ; same is here ; value of drain resistor in input LTP is dependable of exact lower output Jfet Ugs value ; same - value of resistance in paralell to CNY bjt is dependable of exact upper output Jfet Ugs value . so - procedure for making functional Babelfish J2 is something as : 1. test and choose output Jfets ; according to their respective Ugs values - 2. preset input LTP resistance and CNY's parallel resistance 3.preset input LTP CCS at 20mA ; then 4. solder outputs in place and proceed with proper power up procedure #3 demands some tricks how to , with auxiliary PSU ...... or just setting programming resistance in CCS to ~ 0V65/20mA=32R5 more to come - with proper schematic , with parts nomenclature Thanks for the applied science. Let's see if I can dumb it down. So it goes then match some R100's at 2A, find decent matched pair. I'll assume that 1v6 was your vgs. #2 Set to get proper current to CNY pin 1 and across 200r pot. will guess that this pot affects offset. Also, front end Jfets will need drain set with 100R pot (will also affect offset). #3, front end CCS may need to be fiddled with to get 20MA to girly JFETs, too much = white smoke. Does 22r pot help here, or should a pot be in series with 620R? (500r and 200r) This sounds like the making of a Cookbook Someone let me know if I can score good serbian pcb or gerber files. I got JFET's to trade.
Zen Mod Napisano Jun 3, 2010 Autor #251 Napisano Jun 3, 2010 be patient day or two - then we'll have proper schematic with proper nomenclature , so explaining things will certainly be easier ;I'll make small batch of pcbs , so you can count on that ....... but be aware of possibility that someone on big DiyA came with closer approximation of real J2 .I intentionally didn't count parts in J2 , so I use it as general guide , really wanting to Babelfish it , not clone it .edit - all values of pots , along with dedicated series fixed resistors , are chosen to prevent any smoke scenario
sp300b Napisano Jun 5, 2010 #252 Napisano Jun 5, 2010 This seems to be the final revision...Checking still waiting to be done
sp300b Napisano Jun 6, 2010 #254 Napisano Jun 6, 2010 Yummy Manu Hi Manu, Your Pumpkin is completed or not ?
generg Napisano Jun 6, 2010 #255 Napisano Jun 6, 2010 compliment to Zen Mod and friends! It is very pleasant this forum exists and for my purposes its more productive and more open than the big brother at the moment! (everybody building F5) thank you all
Zen Mod Napisano Jun 6, 2010 Autor #256 Napisano Jun 6, 2010 compliment to Zen Mod and friends! It is very pleasant this forum exists and for my purposes its more productive and more open than the big brother at the moment! (everybody building F5) thank you all feel free to contribute whatever and whenever you want
zocky Napisano Jun 7, 2010 #257 Napisano Jun 7, 2010 This seems to be the final revision...Checking still waiting to be done PCB rutirana ručno ili programski?Ne tvrdim da bi se nešto dobilo rutiranjem pod uglom od 45 ali bi se izvestan broj vodova mogao skratiti time.
Zen Mod Napisano Jun 7, 2010 Autor #258 Napisano Jun 7, 2010 ma pusti - da znas kad se u tokudana/noci radi - dobra je i predobra .u svakom slucaju dobra za nas :buehehe:znas i sam da za zadnje cinculacije u svakom programu treba najvise vremena ; evo vec 2 dana ne stizemo da chekiramo vodove
Impuls Napisano Jun 8, 2010 #260 Napisano Jun 8, 2010 Meni licno je ruzna stampa. Moze na njoj da se dokoriguje gomila sitnica. Evo d apocnem da je korigujem:- Q7 okrenuti za 180 stepeni- kond od 4u7 malo odmaci od njega i ti uglaviti R23- da bi se to omogucilo otpornik R35 staviti paralelno R15 zbog estetike a duzina vodova se ne menja- tranzistor levo od Q7 okrenuti za 180 stepeni- R13 i R14 da zamenu mesta a fet izmedju njih zarotirati za 180 stepeni- vod an bazi Q2 moze i bez vie ako se stvi sa gornje strane- na desnoj strani ima jos vise brljotina: vod od desnog kraja R6 ne mora da ide okolo sa dve vie kad moze kratko do fetaeto par korekcija pa posle ispravke nastavljamo
Zen Mod Napisano Jun 8, 2010 Autor #261 Napisano Jun 8, 2010 ja to ne znam - za sve ove godine sam pogubio zivce za pcbs ...... jos iz doba kad se to nije radilo racunarima ....... odnosno - bar ja nisam imao para da kupim frizak 386 i da radim u Tangu kasnije me nije to interesovalo ......... a nadam se i da nece morati dakle - ostaje da Sloba ......... ako ima vremena ............ ne sumnjam da sve primedbe stoje , ali isto tako ne sumnjam da ce i ovako biti 100% funkcionalna ...... a mozda to cak i ja veceras uradim , sto si napisao ......
Impuls Napisano Jun 8, 2010 #262 Napisano Jun 8, 2010 ja za TO92 kuciste koristim komponentu sa malo razvucenim nogicama pa vod od srednje nogice lako prolazi kroz sredinu a ne menja bitno povrsinu za komponentu. A da ce biti funkcionalna i ovako ne znaci i da ce dobro svirati. Gomila nepotrebno isprepletenih vodova samo moze da preuzima smetnje i takve ih pojacava.Na primer imam Hamegove sonde za osciloskop a pored njih imam i kablove sa stipaljkom. Kabel sa stipaljkmo koji nema oklopljen kabel pokupi gomilu smetnji iz prostora i vise nemam informaciju o signalu. Apsurdno je koristiti vrhunske komponente, savrsene topologije a onda sve u***** lose projektovanom stampom. Kod nas u Backoj kazu: sta vredi sto krava daje 25 litara mleka kad se posle ritne i sve prospe.
sp300b Napisano Jun 8, 2010 #264 Napisano Jun 8, 2010 O, Srbijo zemljo, otvori se Ovde ništa ne fali i sve OK, a PCB- u nadjoste 100 mana.
Preporučeni Komentari
Kreiraj nalog ili se prijavi da daš komentar
Potrebno je da budeš član DiyAudio.rs-a da bi ostavio komentar
Kreiraj nalog
Prijavite se za novi nalog na DiyAudio.rs zajednici. Jednostavno je!
Registruj novi nalogPrijavi se
Već imaš nalog? Prijavi se ovde
Prijavi se odmah