ZAPpulse 2.3SE vs. 700XE

Status
Not open for further replies.
Jump to Latest
#221
KLe said:


Hey Gertjan
Are you saying, that the idea came from Lars :whazzat: :cheerful: 😀

Gertjan, do you remember why 4 x RCR, in parallel per rail, are required? Also, any particular reason for choosing 8200uF caps?

Thanks again :wave2:
Click to expand...


I don't know where the original idea came from, I have seen it in several places, just the first place I saw it was on thge LCausio web-site. Maybe Lars can answer your question on who invented it.

There is no reason why you need 4 of them in parrallel, the more the better I would say. Also no reason to limit it to 8200uF, I just had a large stock of 8200uF capacitors. The larger the R value, the better the HF noise suppression as the R in combination with the parasitic series resistance of the caps form a voltage divider, however, with a larger R, the voltage drop increases when suddenly a lot of current is required. This can be solved by using more RCR stages in parallel. With 0.1 ohm for the R, and 4 stages in parrallel, the DC series resistance of the whole thing would be (0.1+0.1)/4 = 0.05 Ohm, which is pretty low. Actually, for suffiently high frequencies, the resistance/impedance is half of that, so 0.025 Ohm (assuming parasitic resistance of the cap is small) as the current will come directly come from the C.

Best regards

Gertjan
 
#222
ghemink said:
I don't know where the original idea came from, I have seen it in several places, just the first place I saw it was on thge LCausio web-site. Maybe Lars can answer your question on who invented it.

There is no reason why you need 4 of them in parrallel, the more the better I would say. Also no reason to limit it to 8200uF, I just had a large stock of 8200uF capacitors. The larger the R value, the better the HF noise suppression as the R in combination with the parasitic series resistance of the caps form a voltage divider, however, with a larger R, the voltage drop increases when suddenly a lot of current is required. This can be solved by using more RCR stages in parallel. With 0.1 ohm for the R, and 4 stages in parrallel, the DC series resistance of the whole thing would be (0.1+0.1)/4 = 0.05 Ohm, which is pretty low. Actually, for suffiently high frequencies, the resistance/impedance is half of that, so 0.025 Ohm (assuming parasitic resistance of the cap is small) as the current will come directly come from the C.

Best regards

Gertjan
Click to expand...

Hi Gertjan
Thankyou, excellent explanation and very much appreciated 🙂 Gertjan, it would appear that the RCR circuit is really for suppressing zener diode noise, ripple and other HF noise. So, it is acting as a HF smoothing circiut between the bridge and the 10000uF cap. Is that correct?

If that is correct, then is it better than placing 4 x 22nF PP caps around the bridge, to suppress zener diode noise?

(PS: Chris, I made a mistake previously, I said .22uF, but, I should have said .022uF caps ... sorry about that ...) :wchair:

Gertjan, what do you think of Carlos's snubber circuit ... 10000uF PS cap// 100nF (MKT) // 0.1R resistor + 3.3nF (MKT) ... for further filtering of airborne junk picked up by the rectifier and compensating for the 10000uF caps inductance? :yes: :emoticon:

Thanks :deer:
 
#225
classd4sure said:
Have a look here:

http://www.lcaudio.com/index.php?page=45

Depending how you look at it, it's either an RCR charging equalizer or a CRC paralled PI filter..... triple duty as a power supply :clown:

At least, that's my guess.
Click to expand...

Thanks Chris
I thought it was 0R1 + 8200uF cap + 0R1 in parallel with and before the 10000uf PS cap. :sorry:

Chris what's a charging equalizer? :bigeyes:

Yes, they are all after the 10000uF PS cap acting like a paralleled CRC network of PI filters??? There are effectively 2 network of paralleled PI filters. 4 x (10KuF + 0R1 + 10kuF) and 4 x (10kuF + 0R1 + 22uF).
Chris, is this the part that the 22uF is playing, although it is quite small?

Wonder, if this could be done with T-Network caps? :drunk:

Thanks
 
#226
hello,

i finished my amp today:



Zappulse1.JPG


5 zappulse 2.3 SE with 1300VA toroidal transformers. It sound very good. i'm very happy😀

Seb
 
#229
ackcheng said:
Sorry guys,

i am a bit lost! Is it CRC or RCR? I thought filtering network is always CRC??

Tanks for your help in clarifying
Click to expand...


OK, I'll try to explain.

First bridge rectifiers (but in my case an SMPS), then for example a first cap, such as 10000uF, then 4 RCR filters in parallel (0.1 Ohm, 8200uF, 0.1 Ohm in my case) and then another 10000uF cap (ELNA cerafine in my case).

So it is more like a C-RCR(4x parallel)-C type of thing.

Best regards

Gertjan
 
#230
KLe said:


Hi Gertjan
Thankyou, excellent explanation and very much appreciated 🙂 Gertjan, it would appear that the RCR circuit is really for suppressing zener diode noise, ripple and other HF noise. So, it is acting as a HF smoothing circiut between the bridge and the 10000uF cap. Is that correct?

If that is correct, then is it better than placing 4 x 22nF PP caps around the bridge, to suppress zener diode noise?

(PS: Chris, I made a mistake previously, I said .22uF, but, I should have said .022uF caps ... sorry about that ...) :wchair:

Gertjan, what do you think of Carlos's snubber circuit ... 10000uF PS cap// 100nF (MKT) // 0.1R resistor + 3.3nF (MKT) ... for further filtering of airborne junk picked up by the rectifier and compensating for the 10000uF caps inductance? :yes: :emoticon:

Thanks :deer:
Click to expand...


The RCR is still another layer of HF filtering. I would still use the snubbers over the diodes as that RCR can not filter out everything because of the series resistance of the caps. So suppressing as much as possible before the RCR is never bad.

I do not have any experience with snubbers on the power supply caps. Depending on what amps you are using I can imagine they could ne beneficial. I think they would not be needed for UcD amps as they have the snubbers already on the amp PCB themselves.

Best regards

Gertjan
 
#232
ghemink said:
OK, I'll try to explain.

First bridge rectifiers (but in my case an SMPS), then for example a first cap, such as 10000uF, then 4 RCR filters in parallel (0.1 Ohm, 8200uF, 0.1 Ohm in my case) and then another 10000uF cap (ELNA cerafine in my case).

So it is more like a C-RCR(4x parallel)-C type of thing.

Best regards
Gertjan
Click to expand...

Thanks Gertjan
I think that you are saying that it is the same as Lars, but, you are using 4 x 8K2uF caps instead of 4 x 10KuF caps and a 10KuF cap instead of the 22uF cap. Gertjan, is that correct?
Lars PS is here http://www.lcaudio.com/index.php?page=45 :cubist:
The same diagram is shown in the next post that Chris has posted. :emoticon:


ghemink said:
The RCR is still another layer of HF filtering. I would still use the snubbers over the diodes as that RCR can not filter out everything because of the series resistance of the caps. So suppressing as much as possible before the RCR is never bad.
Click to expand...
Yes, I think that you are right. The rectifier caps make a huge difference in controlling zener diode noise. Having said that, I think the FRED bridges that Chris is using may not need the recitifier caps? :Pirate:


ghemink said:
I do not have any experience with snubbers on the power supply caps. Depending on what amps you are using I can imagine they could ne beneficial. I think they would not be needed for UcD amps as they have the snubbers already on the amp PCB themselves.
Click to expand...
Yes, I think that you are right, again. I have the Zap2.3SE amps, not UcDs, and Lars is not recommending this approach, as such, but I believe that PS snubbers could be beneficial, also. Can there be too much HF damping? :usd:
 
#233
KLe said:


Thanks Gertjan
I think that you are saying that it is the same as Lars, but, you are using 4 x 8K2uF caps instead of 4 x 10KuF caps and a 10KuF cap instead of the 22uF cap. Gertjan, is that correct?
Lars PS is here http://www.lcaudio.com/index.php?page=45 :cubist:
The same diagram is shown in the next post that Chris has posted. :emoticon:



Yes, I think that you are right. The rectifier caps make a huge difference in controlling zener diode noise. Having said that, I think the FRED bridges that Chris is using may not need the recitifier caps? :Pirate:



Yes, I think that you are right, again. I have the Zap2.3SE amps, not UcDs, and Lars is not recommending this approach, as such, but I believe that PS snubbers could be beneficial, also. Can there be too much HF damping? :usd:
Click to expand...


You are right about my setup being almost the same as the one of Lars.

I don't think one can have too much HF damping, depends of course how it is achieved. I have my reservations when it is done with LC networks because of possible resonances as said earlier. That's why I limit myself to RC damping (although the UcD modules have a few small L's onboard for HF filtering.

Best regards

Gertjan
 
#234
ghemink said:
You are right about my setup being almost the same as the one of Lars.

I don't think one can have too much HF damping, depends of course how it is achieved. I have my reservations when it is done with LC networks because of possible resonances as said earlier. That's why I limit myself to RC damping (although the UcD modules have a few small L's onboard for HF filtering.

Best regards

Gertjan
Click to expand...

Hi Gertjan
Do you believe that this is the best way to do your power supply caps? :tons:

I like this explanation that Lars gave, as it makes sense to me, "This is what i have (maybe wrong word) called current woppling ... If you have say 5 capacitors in parallel, in a dual rail connection, where the rectifier is in one end, and the amplifier is in the other, and we decide the ESR of each capacitor is 10 milliOhms, and the connection resistance between each capacitor is 2 milliOhms. The capacitor nearest to the rectifier is charged with the largest current (since it has only 10 mOhm ESR) , and the one furthest away is charged with only about half the current (since it has 20 mOhm as seen from the rectifier). Then when the charging spike is over, the charge rolls forward from the first capacitor to the last and attempts to equalize the charge over the entire bank. So each capacitor ends up having the same voltage. The resistors ensure that the charging spike is equally distributed over the entire capacitor bank, and also equally discharged from all 4 capacitors to the amplifier". :checked:

I really like the BHC T-Network caps, so, I will see if I can use these in this arrangement? :hypno2:

Yes, wrsto HF damping you are probably correct. I will have a think about it? :tilt:

Thanks again
 
#235
KLe said:


Hi Gertjan
Do you believe that this is the best way to do your power supply caps? :tons:

I like this explanation that Lars gave, as it makes sense to me, "This is what i have (maybe wrong word) called current woppling ... If you have say 5 capacitors in parallel, in a dual rail connection, where the rectifier is in one end, and the amplifier is in the other, and we decide the ESR of each capacitor is 10 milliOhms, and the connection resistance between each capacitor is 2 milliOhms. The capacitor nearest to the rectifier is charged with the largest current (since it has only 10 mOhm ESR) , and the one furthest away is charged with only about half the current (since it has 20 mOhm as seen from the rectifier). Then when the charging spike is over, the charge rolls forward from the first capacitor to the last and attempts to equalize the charge over the entire bank. So each capacitor ends up having the same voltage. The resistors ensure that the charging spike is equally distributed over the entire capacitor bank, and also equally discharged from all 4 capacitors to the amplifier". :checked:

I really like the BHC T-Network caps, so, I will see if I can use these in this arrangement? :hypno2:

Yes, wrsto HF damping you are probably correct. I will have a think about it? :tilt:

Thanks
Click to expand...


I dont' know about that.

What you need is bandwidth, probably even more so in class d! You dont' want resonance, but you sure don't want it overdamped. Critically damped, if required.

I think if you're using T-networks and they're wired properly you don't need to worry about this sort of thing because you're filtering is built in already.

KLe the rest of that thread also held some very informed arguments to the contrary of the usefulness of such a circuit, and a point I found interesting in particular, is this "woppling" thing such a problem to begin with?

Would you appreciate a diagram on how I'm going to parallel my 4 poles? I'd have to do it in mspaint or find a scanner so if you're not interested in doing it anymore I'll save myself the time.

Regards,
Chris
 
#236
classd4sure said:
I dont' know about that.

What you need is bandwidth, probably even more so in class d! You dont' want resonance, but you sure don't want it overdamped. Critically damped, if required.

I think if you're using T-networks and they're wired properly you don't need to worry about this sort of thing because you're filtering is built in already.

KLe the rest of that thread also held some very informed arguments to the contrary of the usefulness of such a circuit, and a point I found interesting in particular, is this "woppling" thing such a problem to begin with?

Would you appreciate a diagram on how I'm going to parallel my 4 poles? I'd have to do it in mspaint or find a scanner so if you're not interested in doing it anymore I'll save myself the time.

Regards,
Chris
Click to expand...

Hey Chris
Good to hear from you ...

Yes pleeeease ... I would very much appreciate a diagram on how you are going to parallel your Jensen 4 poles? :drink: You bet ya!!!

Actually, the Virtual 4 pole PS looks like it requires much more room than parralleling the T-Network caps? :cubehead:

Thanks again
 
#237
classd4sure said:



I dont' know about that.

What you need is bandwidth, probably even more so in class d! You dont' want resonance, but you sure don't want it overdamped. Critically damped, if required.

I think if you're using T-networks and they're wired properly you don't need to worry about this sort of thing because you're filtering is built in already.

KLe the rest of that thread also held some very informed arguments to the contrary of the usefulness of such a circuit, and a point I found interesting in particular, is this "woppling" thing such a problem to begin with?

Would you appreciate a diagram on how I'm going to parallel my 4 poles? I'd have to do it in mspaint or find a scanner so if you're not interested in doing it anymore I'll save myself the time.

Regards,
Chris
Click to expand...



Hi Chris,

I agree, if you use T network caps or 4-pole caps, then I would also say that you don't need this RCR stuff. The RCR stuff is basically a poor man's 4-pole cap (or T cap), I guess that is why Lars called it a virtual 4 pole network. I had lots of these caps (the 8200uF 63V Nichicons) and wanted to use them. The caps close to the amps, so the caps after the RCR are high quality caps (ELNA Cerafine, these are pretty low ESR) and I think those caps would determine the sound quality of the amp as the other caps are connected via the small Rs and thus less directly coupled to the power rails of the amp.

Best regards

Gertjan
 
#238
ghemink said:




Hi Chris,

I agree, if you use T network caps or 4-pole caps, then I would also say that you don't need this RCR stuff. The RCR stuff is basically a poor man's 4-pole cap (or T cap), I guess that is why Lars called it a virtual 4 pole network. I had lots of these caps (the 8200uF 63V Nichicons) and wanted to use them. The caps close to the amps, so the caps after the RCR are high quality caps (ELNA Cerafine, these are pretty low ESR) and I think those caps would determine the sound quality of the amp as the other caps are connected via the small Rs and thus less directly coupled to the power rails of the amp.

Best regards

Gertjan
Click to expand...


Hello, I think you're right with that too about what influences the sound, and I knew you'd have your reasons for doing it 🙂 Your work and projects are amongst the very best in my book.

I've been wondering, and you've probably tried this. A CRLRC where the R's are bleeders and damp resonance?

Best,
Chris
 
#239
classd4sure said:



Hello, I think you're right with that too about what influences the sound, and I knew you'd have your reasons for doing it 🙂 Your work and projects are amongst the very best in my book.

I've been wondering, and you've probably tried this. A CRLRC where the R's are bleeders and damp resonance?

Best,
Chris
Click to expand...


Hi Chris,

I have tried some simulations with SwitchCad in the past with LRC like combinations. However, I could not get a solution that I liked. I do not remember the details anymore but I did not like the response of the network on sudden current changes, adding more R to damp the resonances increases voltage drops which I did not like either. However, everything was based on simulations, maybe in reality it can work??? probably it is best used in Class A amps and tube amps?

Best regards

Gertjan
 
#240
ghemink said:
Hi Chris,
I agree, if you use T network caps or 4-pole caps, then I would also say that you don't need this RCR stuff. The RCR stuff is basically a poor man's 4-pole cap (or T cap), I guess that is why Lars called it a virtual 4 pole network. The caps close to the amps, the caps after the RCR, are high quality caps (ELNA Cerafine, these are pretty low ESR) and I think those caps would determine the sound quality of the amp as the other caps are connected via the small Rs and thus less directly coupled to the power rails of the amp.

I have tried some simulations with SwitchCad in the past with LRC like combinations. However, I could not get a solution that I liked. I do not remember the details anymore but I did not like the response of the network on sudden current changes, adding more R to damp the resonances increases voltage drops which I did not like either. However, everything was based on simulations, maybe in reality it can work??? probably it is best used in Class A amps and tube amps?

Best regards

Gertjan
Click to expand...

Hi Chris/Gertjan

Yes, a poor man's 4-pole cap (or T cap). :bigeyes: What you are both saying makes complete sense. The T-Network caps are very impressive, very revealing. :hot: Parrallelling them should lower PS output impedance, which in itself should make the PS faster and cleaner, and the PSR should also be better. Guys, what are your thoughts on all the benefits of parallelling the Jensen 4 poles? 😀

Chris, to reiterate ... you bet ya, I would very much appreciate a diagram on how you are going to parallel your Jensen 4 poles?
:hypno1: :Popworm: :whazzat: :captain: :judge:

Thanks again, guys
 
Status
Not open for further replies.
Top Bottom