Like these
http://FRH10A15.pdf FCH10A15.pdf
and KCH20A10.pdf
Regards
James
An externally hosted image should be here but it was not working when we last tested it.
http://FRH10A15.pdf FCH10A15.pdf
and KCH20A10.pdf
Regards
James
Poobah, Thanks. Yes, the film caps would have been one of the reasons I would not expect to see the hash or noise survive. I ussually use some big E-Caps bypassed with Films going onto the amp board also...
Would using a small value inductor like 50uH or so after my 6mH inductor take care of the hash??? Actually, I have a monster 25uH unit I scavenged from a UPS...
Would'nt the Film caps also be benificial on the bridge side of the Pi filter? +V>>GND. Just more attenuation of the hi freqs???
Beads Hmm, I can try some of those too...
When I get this new Supply done I'm going to play some games with the electric drill test... Have you seen that here some time ago? Funniest Test I ever heard... Take an old electric drill and plug it into the same outlet as your supply... hit the trigger and look around with the scope... I'm planning on using a common mode Pi filter on the input also... Could be interesting... I'm going to try it without all the hi freq tricks and then start adding Films and such. I want to see just how much of a problem and benifit all this stuff might be...
I guess just to be true to the thread I'm going to go check out your UF series parts..., Thanks again
Would using a small value inductor like 50uH or so after my 6mH inductor take care of the hash??? Actually, I have a monster 25uH unit I scavenged from a UPS...
Would'nt the Film caps also be benificial on the bridge side of the Pi filter? +V>>GND. Just more attenuation of the hi freqs???
Beads Hmm, I can try some of those too...
When I get this new Supply done I'm going to play some games with the electric drill test... Have you seen that here some time ago? Funniest Test I ever heard... Take an old electric drill and plug it into the same outlet as your supply... hit the trigger and look around with the scope... I'm planning on using a common mode Pi filter on the input also... Could be interesting... I'm going to try it without all the hi freq tricks and then start adding Films and such. I want to see just how much of a problem and benifit all this stuff might be...
I guess just to be true to the thread I'm going to go check out your UF series parts..., Thanks again
The inductor you speak of is called a "hash choke". I would put in the downstream end of the supply as well as any film caps (but you probably don't need it).
Ferrite beads would work best at the power input of your amp board.
And yes that drill test would be most effective... may sure to have someone else do the test far away from you amp... otherwise you may measure radiated, rather than conducted, EMI.
Try an AM radio tuned to an empty spot to see how radiated noise a drill puts out.
You have to calculate the energy losses in film caps before you put them just anywhere... close to the bridge may not be so good... just a guess.
Oh... you're most welcome... be sure to tell us how it works out!
Ferrite beads would work best at the power input of your amp board.
And yes that drill test would be most effective... may sure to have someone else do the test far away from you amp... otherwise you may measure radiated, rather than conducted, EMI.
Try an AM radio tuned to an empty spot to see how radiated noise a drill puts out.
You have to calculate the energy losses in film caps before you put them just anywhere... close to the bridge may not be so good... just a guess.
Oh... you're most welcome... be sure to tell us how it works out!
Thanks Poobah & Flq for ur replies.
But I am still confused. You mean to say that it is the fastest among all, leaving aside capicitance, reverse leakage or forward drop. I have used UF series as a snubber and it works well, but not in the config as shown in pic, where they recommend schotky.
Why so?
ajazz
But I am still confused. You mean to say that it is the fastest among all, leaving aside capicitance, reverse leakage or forward drop. I have used UF series as a snubber and it works well, but not in the config as shown in pic, where they recommend schotky.
Why so?
ajazz
Attachments
I assume this a class d amp? In this circuit, these diodes are called "freewheeling diodes" as they allow "buck" action. Schottkys are chosen for their efficiency and their lack of recovery effects. In this circuit the transistors see the recovery current, large recovery currents can develop voltage across the source inductance and cause difficulties in maintaining good gate drive.
Wow, I was only thinking of the age old linear AC supply. "Big Iron" type of thing. So my Class A designs have a good source and the subdleties I'm chasing are in the Class A circuit design and components.
I don't have much against "Digital", generally speaking. But, as I have plenty of other digitally controlled things in my life and at work, that are not worth playing with myself anymore, I enjoy chasing the performance available with "classic" analog music reproduction. I'd still be using 2-track tape and turntables if I could. Why dosent someone make an analog CD??? Like scatchless, noiseless Vinyl!!! I don't need 400 songs on every disk...
I built myself a custom H.D.. No electric start, it only kicks, no electronic ignition, it's a magnito, no battery at all, no instuments, no front fender, no blinkers, no lugage rack, no fairing, no radio. It even has a 65 year old front end. But, it gets all kind of looks and it is so rewarding to ride and own. Your classic 50 year old Hot Rod Bike! And like a simple Nelson Pass Design, It's an unbelieveable performer...
Ajazz, (I am also assuming Class D???) try some class A, you'll like it
Just a thought ...
I guess I'll get some sh#$%^ for that haa??? O.K. I think the deal is,,, these diodes are directly related to converting the 1s and 0s to filtered analog music. You can't easily deal with the switching spikes and recovery problems associated with using the wrong type of diode... The other thing might be like in N.P.s latest commercial venture. 1st Watt... Although small, the 0 crossing, or low current noise perfomance of Schotky's is better than other types.
Just another thought
I don't have much against "Digital", generally speaking. But, as I have plenty of other digitally controlled things in my life and at work, that are not worth playing with myself anymore, I enjoy chasing the performance available with "classic" analog music reproduction. I'd still be using 2-track tape and turntables if I could. Why dosent someone make an analog CD??? Like scatchless, noiseless Vinyl!!! I don't need 400 songs on every disk...
I built myself a custom H.D.. No electric start, it only kicks, no electronic ignition, it's a magnito, no battery at all, no instuments, no front fender, no blinkers, no lugage rack, no fairing, no radio. It even has a 65 year old front end. But, it gets all kind of looks and it is so rewarding to ride and own. Your classic 50 year old Hot Rod Bike! And like a simple Nelson Pass Design, It's an unbelieveable performer...
Ajazz, (I am also assuming Class D???) try some class A, you'll like it
Just a thought ...
I guess I'll get some sh#$%^ for that haa??? O.K. I think the deal is,,, these diodes are directly related to converting the 1s and 0s to filtered analog music. You can't easily deal with the switching spikes and recovery problems associated with using the wrong type of diode... The other thing might be like in N.P.s latest commercial venture. 1st Watt... Although small, the 0 crossing, or low current noise perfomance of Schotky's is better than other types.
Just another thought
Thanks Poobah & Flq.
If Schotky diodes are so good, which can be used everywhere then why are they still manufacturing Ultra, hyper & fred diodes.
Is cost the only factor. I don't mind keeping a single inventory rather keep all varieties.
Dear flq though I love ClassA and have made many varients for personel use but it can't surpass the sound quality of Class D at high power levels (good design). Well thats my experience and feel that a good D amp is the future of sound.
BTW. So I should stick with Schotky - - thats right ?
regards
ajazz
If Schotky diodes are so good, which can be used everywhere then why are they still manufacturing Ultra, hyper & fred diodes.
Is cost the only factor. I don't mind keeping a single inventory rather keep all varieties.
Dear flq though I love ClassA and have made many varients for personel use but it can't surpass the sound quality of Class D at high power levels (good design). Well thats my experience and feel that a good D amp is the future of sound.
BTW. So I should stick with Schotky - - thats right ?
regards
ajazz
Hell, I don't know... I work with off line SMPs in the daytime and we care about that recovery stuff and all mostly on the primary side. but we are jamming square waves on and off of an inductor at high voltage and 100kHz+. In the output, there are other considerations but schotky's are popular there if the turns ratio allows it(reverse voltage thing ussually). We dont ussually care about the parameters we've been talking about here in the SMPs because of the harsh invirenment. Noisy as that electric drill test without the drill.
As an example; When you have 5-10 amps dc output current the schotky forward voltage being so low provides substantial effeciancy increase.ie... 5A X 1V =5watts lost in that "Freewheeling" diode if it is a fast or ultra fast component. If it is a 5V supply you just lost 20% of your power in that diode!And at that level it will be expensive and need a heat sink. However, 5A X .5V is only 2.5watts maybe in a schotky. about 10% loss!... That is a big deal... Schotky's are expensive in the SMPs world but those people build many thousands of units so they scrimp werever they can... We use these things a few at a time... It is not acttually that simple but I'm sure this belongs in another forum catagory if I continue...
Basically the lower current operation of any diode has shot noise. Even the LEDs we sometimes use for bias ref voltage... The schotkty not being your normal silicon PN junction has lower noise in that region than the others... To us audio finatics that is a big deal. In my power supply, the diode will see the 0 crossing portion of the AC waveform 60 or 120 or whatever times a second, and that is the noisy area...
For my 23-24VDC Circlotron supplies I'm going to give it a try... I'm a little concerned about getting it turned on with that pile of Cap after it though...
As an example; When you have 5-10 amps dc output current the schotky forward voltage being so low provides substantial effeciancy increase.ie... 5A X 1V =5watts lost in that "Freewheeling" diode if it is a fast or ultra fast component. If it is a 5V supply you just lost 20% of your power in that diode!And at that level it will be expensive and need a heat sink. However, 5A X .5V is only 2.5watts maybe in a schotky. about 10% loss!... That is a big deal... Schotky's are expensive in the SMPs world but those people build many thousands of units so they scrimp werever they can... We use these things a few at a time... It is not acttually that simple but I'm sure this belongs in another forum catagory if I continue...
Basically the lower current operation of any diode has shot noise. Even the LEDs we sometimes use for bias ref voltage... The schotkty not being your normal silicon PN junction has lower noise in that region than the others... To us audio finatics that is a big deal. In my power supply, the diode will see the 0 crossing portion of the AC waveform 60 or 120 or whatever times a second, and that is the noisy area...
For my 23-24VDC Circlotron supplies I'm going to give it a try... I'm a little concerned about getting it turned on with that pile of Cap after it though...
freakyone said:
Hi Freakyone
Try enquiries@wescomponents.com
They have dual schottky rectifiers of both polarities
S16C60A & S16C60C 16 Amps @ 60 Volts <AUS$2 each
and S30D60A & S30D60C 30 Amps @ 60 Volts <AUS$3 each
Cheers
Rob
True enough freakyone, do keep in mind that MOV's are consumable... they can only take so many whacks.
My point is that Schottkys fail because of a purely voltage related mechanism.
Silicon PN diodes will avalanche (zener-like) when their ratings are exceeded... the failure mechanism then is about power and the thermal capacity of the die. PN's are just tougher.
My point is that Schottkys fail because of a purely voltage related mechanism.
Silicon PN diodes will avalanche (zener-like) when their ratings are exceeded... the failure mechanism then is about power and the thermal capacity of the die. PN's are just tougher.
Well, as long as we are still on the subject. Let me ask something mostly off thread and very un-Zen-like but, a relative idea... How easy would it be to design a synchronuos rectifier circuit that would solve all these problems, noise, reliability etc. and be more efficient. Maybe include a soft start??? Is a sychronuos bridge a viable solution???
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.