Polypropylene vs electrolytic in Power supply

[duderduderini]

Hi Guys.
Assuming I can get ppropylene caps of say 10uF would these be beter for filtering in a ps than electrolytic?
I have looked at the specs and the pp's have lower esr, higher ripple current and dont dry out .. the v rating is also 1000Volts.
i ran a sim in duncans psu designer and 10uF in a clc and then an rc seems to look alright.
Anyone have any advice..
They are expensive but given you dont need to series them up and then use resistors etc etc I woonder if it really that expensive.
It seems to me that pp have few disadvantages
Thanks for anyones help
Nick

[sorenj07]

Metal film capacitors of any kind are superior in performance in every way to electrolytics. It just depends on whether that superiority is appreciated in its use. I use metal film capacitors in my PSU and don't mind the way my amp sounds but how much of that "goodness" is due to lower ESR is anyone's guess. Maybe if I had a scope I could do some real measurements. Rule of thumb: If you have the room to use metal film, do so

One more thing: polypropylene or any other MF caps tend to self-heal, in other words if some electricity punches through the dielectric, it just vaporizes the metal that it reaches, and leaves the rest intact. Not an excuse to "over-volt" the caps but a nice improvement over electrolytic caps exploding or whatever.

[duderduderini]

Hi The caps i have tracked down have a rating of 1000V given the volts out of the trans will be max 550V i think this margin is good enough (5u4g into clc)
Looks like I will bite the bullet and get some.....
What values do you think are adequete for psu filtration.
I found that 10U was noticeably better looking for ripple than say 4uF
Thanks
Nick

[sorenj07]

Keep in mind that the input capacitor in a CLC filter is a very critical part. The higher the value in uF this capacitor is, a: the less well the PSU is regulated, b: the higher the B+ voltage will be (to a limit), and c: the more stressed the rectifier will be. Read some 5U4 datasheets to get an idea of some implementations. I don't know what the demands of your schematic are off the top of my head but I suggest scratching the first cap altogether if you can afford the drop in B+. However, your choke must be of high enough inductance in order for the choke-input power supply to work, usually 10H is a good starting point.

[duderduderini]

Hi
I have a 10mH choke.. when i sim the psu on duncans psu designer, the b+ is down which isnt so bad but the ripple seems more than a traditional clc topology
what value cap do you deploy just after the choke? Would you plae an rc filter as well
Thanks

[duderduderini]

Dave I got your email and replied but it bounced and your email address is not taking emails on diyaudio so I am pasting what i sent
Hi Dave
I have found one here is the link
http://www.mouser.com/search/Product...98-UNL10W10K-F
I have attached the pdf fo your perusal.
The only trouble is the cost.. single unit price is 28 dollars US for 1 or 17.20 usd if you buy 50 (which I am thinking to do)
I like the specs.. for a start it v ratig is 1000 Volts its current rating is 17A whereas even the best electrolyitic cap has a rating of only 1A at the same frequency. They sound better they are more durable (they heal if the film is punctured).
I have dealt with Mouser and they are very good.. stuff takes round 2 weeks to get here.
Now I am into a bit of stockpiling as i bel;ieve that sooner or later these High Voltage caps will be in short supply. If you compare the unit price with say the 50 unit price its 10 dollars cheaper and the 100 buy price is almost half!!!
If you want to entertain a bulk buy let me know we can do some kind of agreement etc.
I hope this helps
I have done the power supply sim with duncans psu and 10uF is ok for the ripple but it needs 2 maybe 3 stages ( one inductor and 2 rc stages) alternatley I might look into say a ppropylene input cap stage then after the inductor a JJ multisection cap(round 30 dollars aud).. The enticing thing about the metallised ppropylene option is the 1kV rating and the amps which is easily 17 times better than the best electolytic
regards
Nick Magelakis

[AndrewT]

Hi,
do you need 1000Vdc rating?

Motor start PP caps are cheap upto about 18uF, going upto 40uF gets more expensive.

these are available for 240Vac supplies and are usually rated to about 400Vdc.
I think some are available for 400Vac (~600Vdc) but I have no idea on price or availability.

[duderduderini]

Hi Andrew
Thanks for the tip.. Ok Ok you have spotted the Obsessive nature in me. I used to wear my seat belt whilst watchhing a movie at the drive in .. cant be too carefull.
I suppose i like to over engineer cause I figure if you run something at 50% of its load its less likely to fail than something at 80% still motor start caps in parallel would give me the margin (or close to)
Regards
Nick

[AndrewT]

Originally posted by duderduderini
what value cap do you deploy just after the choke? Would you plae an rc filter as well

Hi,
keep your best sounding smoothing cap for the last in the string CLCLC or whatever you choose.

The first will have least effect on the sound quality.

I just looked up Rapid and they stock 450Vac continuous, they should be good for 600Vdc. These are Ducati, if you want to look up a datasheet for DC and other info.
18uF =£4.05 in single, 10uF is only £2.77, falling to £2.53 for 10 off.
two channel CLC needs 4off, pre-amp another 4off, now what could you use the last two for?
Ah! across the mains as suggested in another thread (J.Curl?)

[ianc13]

If you want something a bit bigger - these are 80uF.

$12 a pop. I am using some in a project that I am building at the moment, 500VAC and the seller reckons around 850VDC. I'm using them around 600VDC.

The only issue is size - they are huge - which makes fitting them inside my project a bit interesting.

80uF motor runs