Optimal supply design for UCD and Zappulse modules

Status
Not open for further replies.
Jump to Latest
#81
Golgoth said:

- UcD400 can output 210W into 8 ohms, let's round this to 256W so it's an integer number of doublings (we'll be wrong by less than 3dB anyway), namely 8 doublings wrt 1W (256=2^8):
-> 8 times +3dB = +24dB -> 88+24=112 dB SPL @1m

- two doublings of distance for 4m (1*2*2=4):
-> 2 times -6dB -> -12dB -> 112dB -12dB = 100dB SPL @4m

-two enclosures:
-> +3dB -> 100dB +3dB = 103dB SPL @4m for 2 enclosures
Click to expand...


OK, back to the power supply. That means that the power of the UcD is 256Watt maximum (8Ohm), which means 512 VA toroid per channel, right? Since it is only 210Watt 500VA should suffice.
And the caps don't change, still 2*6250uF per channel. And that gives 103dB with to channels at 4 meter in hifi.
 
#82
Just to interrupt you guys again with a cap vendor. 😉
How about this one:

http://www.kendeil.com/index.htm
Personally I would be interrested in 33000uF/75V for my Zappulse modules (two per channel). The specs say ERS is 11 milliOhm at 100 Hz and 20 degrees C. Is that good?

I believe they are also used in Graaf amps.

They are not cheap, here you can find some prices: http://www.herweijerelectronics.nl/shop/caps/elco.htm

Below it you see a pic of Mundorf M-Lytic. I have no specs of them, but they are probably the same as these krummer:
http://www.krummer-kondensatoren.de/
 
#83
I want to take it form the other side:

I now have a 500VA toroid in my amplifier. Wth this it can delver 250W to the UcD's. That is 125W per UcD.
125W are 7 doublings, that's 21dB plus, makes 88+21=109dB @1m. At 4m that is 109-12=97dB. Two speakers makes 97+3=100dB. That should suffice for a living room?!

In this amplifier there are 5 Ucd's to drive 5 speakers in case of a 5-channel DVD. Assuming that the speakers are the same, we get 50W (250/5) per channel. That are 5 to 6 doublings, so about 16dB+, makes 104dB. at 4m that is 104-12=92dB. But there are 5 speakers, so 92+(4*3)=104dB. Should be enough for my living room.

What I do not understand is what does the capacity. Why do youn need how much (for this situation).

Frans
 
#85
FransDHT said:
I want to take it form the other side:

I now have a 500VA toroid in my amplifier. Wth this it can delver 250W to the UcD's. That is 125W per UcD.
125W are 7 doublings, that's 21dB plus, makes 88+21=109dB @1m. At 4m that is 109-12=97dB. Two speakers makes 97+3=100dB. That should suffice for a living room?!

In this amplifier there are 5 Ucd's to drive 5 speakers in case of a 5-channel DVD. Assuming that the speakers are the same, we get 50W (250/5) per channel. That are 5 to 6 doublings, so about 16dB+, makes 104dB. at 4m that is 104-12=92dB. But there are 5 speakers, so 92+(4*3)=104dB. Should be enough for my living room.

What I do not understand is what does the capacity. Why do youn need how much (for this situation).

Frans
Click to expand...

Your maths look ok, except 5 speakers is a bit more than 2 doublings, not 4 !

For capacitance, since you will be using your modules at 50/400 = 1 /8 of the power, you can divide what was computed previously by 8 obviously, cf Bruno's original rule what matters is actual power drawn from the supply.
 
#87
rail caps

Hi,

after all discussions I'm still not sure, what's best, lowest or moderate ESR. I'm pretty sure, that a high allowable ripple current cannot be bad, but this is possible only by achieving lowest ESR. Which may introduce high recharging current spikes.
If I'd go for lowest ESR in conjunction with high ripple current without any price restrictions, I would use Epcos Sikorel or Vishay PED-ST series. The Sikorel caps exhibit outstanding reliability data additionally.

Unfortunately I do not find the source, where to be read, that Aerovox caps exhibit a modest ESR only, around 30mOhms (in the interesting cap/voltage range), the ripple current was accordingly low.

The ESL question should be answered already by the higher wiring inductance (two parallel wires, approximately):
L{nH}=2*l*[ln(2*a/d)+µr/4],

where l = length of the wire, a = distance between the wires, d = their diameter (all in cm), µr = their relative permeability,
that gives in my setup for the two rail wires "in parallel":

L{nH}=2*20cm*[ln(2*1cm/0.18cm)+1/4]
L=106nH for each wire.

That is 10 times the cap ESL, minimum. And the wiring inductance may be of a high Q, possibly causing some ringing, if stimulated by current spikes.

Somebody stated this in earlier discussions already: so called Audio grade components do not need to have exceptional performance necessarily, marketing experts sell them by using "euphonious" brands and looking a million dollars.

Timo
God and RF are cutting their own path.
 
#88
Golgoth said:

For capacitance, since you will be using your modules at 50/400 = 1 /8 of the power, you can divide what was computed previously by 8 obviously, cf Bruno's original rule what matters is actual power drawn from the supply.
Click to expand...


FransDHT said:

That means that the 500VA and the 2x10.000uF should be enough. Jan Peter, you are right after all! I shouldn't worry so much.
Click to expand...


Ah yes my mistake: previous computation was 50000/8= 6250µF, times 5 channels around 30000uF per rail corresponding to 210W per channel (full power on 8 ohms). You'll be using about 1/4 of that (50W/channel) so 30000/4 = 7000uF would be enough, so 10000uF, and not 10.000 uF :clown: seems just fine.
 
#89
RIFA caps?

Hi.

Have any of you tried Rifa caps? I have real trouble finding large powersupply caps for my UcD400 monoblock project, but the Rifas I can buy online from Sweden:

http://www.elfa.se/elfa-bin/dyndok.pl?vat=0&dok=2012735.htm

Are they any good? Also, I wonder if I should get the 15000uF/100V ones or the 22000uF/100V. Is 15000 uF enough, is 22000 uF better or maybe even worse?

Thanks

(I'll take some pictures of my project when it's done and post them somewhere of course)
 
#90
Dual mono Zappulse 2.2SE

Hi All,

I have received my second Zappulse 2.2SE (with BG capacitors to match the first module) and would like to build a dual mono amplifiers in one or two 2U rack enclosure(s). The amplifiers would be used to drive my home made ESL's. The electrostatics present rather difficult load to drive with impedance that drops below 1ohm above 16kHz.

I did try a quick setup with 2 x 40V (AC) 300VA toroid, standard 35A bridge and 4 x 8000mf /75V Lelon (?) electrolytics and it worked nicely. It actually outperformed a couple of reputable amplifiers quite easily.

To ensure I don't make silly errors and to exploit the full module potential I would appreciate your comments in regard to the following:

1. Should I put both modules in one 2U rack case or have each module in its own case?

2. If both modules are in one case, should I have two or one transformer?

3. What should be the transformer(s) VA rating (within reason)?

4. Are the two Rifa PEH200 electrolytics 22000uf/63v per rail/module sufficient and acceptable for 2x40V (AC) secondary windings? These capacitors are bulky, have reasonable low ESR and most importantly are localy available from RS Compononents.

5. Should I use standard 35A bridges with snubbers or go for something like MUR3020WT bridge?

At the moment I have one LC Audio soft start module.


Thank You,
Zvon
 

Attachments

  • module1s.jpg
    module1s.jpg
    33.1 KB · Views: 1,023
#91
One more question (probably a real clueless one...):

I have a transformer with a center-tapped secondary, ie. there are three output wires, 40-0-40 VAC. I have enough rectifier diodes to do separate rectification of the positive and negative rails.

Is that possible to do with a center tapped transformer only with one that has two independent secondary windings? If yes, how do I connect the rectifier bridges?
 
#95
That looks excellent! I should think that it would work very well for the UCD400 as well?

Fancy making up a job lot of those? I would be in the market for a pair (I would want to run dual mono in my next setup, but being able to put 20,000uF per channel would be cool).

Ed
 
#96
Hi Ed,

You're right, with this PSU you can drive a pair of UCD180 (2x10000µF63V per UCD) or one UCD400 (4X10000µF/80V).

I will check and test this design at the begining of january when UCD180 will be available.
Then, some pcb's will be available for peope who are intersted about.

I'm desinging a new PSU (in progress) to drive a pair of UCD400. I'm plane to use 8X10000µF/80V (4X10000µF for each UCD400).

This design will be available very soon.

😉
Franck
 
#99
Re: capacity:

maxlorenz said:
What is the absolute minimum capacity for an UCD400 stereo amp PSU??
I think maybe it's better to have very good quality PS caps with reasonable value than very big, lower quality PSU. When I tried my UCD180 amp with 4700uF*2 caps it sounded very fast and punchy, with easy going bass. I augmented the capacity anyway 😀

What do you think, guys? 😕
Click to expand...


I think this is all covered in this thread and the UCD180 thread.

Fewer better quality caps is recommended by Bruno. However, there is a suggested formula earlier in this thread which basically suggests that the cap sizes are determined by speaker load rather than amp wattage.
 
Status
Not open for further replies.
Top Bottom