LC ZAPpulse Amplifier Q & A

[Rob Fens]

Rifa caps

Hey,

It is well known in Denmark, that Zappulse and Rifa caps don`t make a good partnership.

Hilsen Jan Jensen

Dear Hilsen Jan,

At the moment i use the blue computergrade Spragues 40V 40.000mF 4 pieces in my ZAP 2.2SE supply. The supply voltage is +- 36V I like the sound of the Zap very much (see my customer feedback on the LC site). Since i want more power, i bought a different transformer and RIFA caps PEH169MU533oQ (63V 33.000mF 150degrees Celcius). The supply voltage wille be +- 60V. In your message i read that Zaps and Rifas don't make a good partnership. Both products are really high end, so can you please give me some details about the reason why they don't suit eachother before i spend too much time and money on something worse than i have now.

Best regards,
Rob Fens.

 

[Lars Clausen]

I mostly played with PEH200's (the 85 C series), your PEH169's might well be better.

With PEH200's the mid and top gets all blurry compared to for example Nippon Chemocon's. If used with a Class A/B amplifier the RIFA's perform great. So it has to do with sonic properties of ZAPpulse together with RIFA caps. I could improve the performance with use of Black Gate VK's in parallel with the RIFA's.

 

[ghemink]

I mostly played with PEH200's (the 85 C series), your PEH169's might well be better.

With PEH200's the mid and top gets all blurry compared to for example Nippon Chemocon's. If used with a Class A/B amplifier the RIFA's perform great. So it has to do with sonic properties of ZAPpulse together with RIFA caps. I could improve the performance with use of Black Gate VK's in parallel with the RIFA's.

Are Nippon Chemicons really that good? I have loads of them (8200uF 63V) as I happen to live in Nippon. Plan to use them with a virtual 4-pole type of setup, something like your virtual 4-pole supply and driven by an SMPS instead of a conventional supply.

Best regards

Gertjan

 

[Lars Clausen]

ghemik: I think they have a very clear, but still relaxed sound to them. Compared with other 'standard grade' brands, i find them to be the best i have tried so far. And IMO better than also some socalled 'high grade' brands like BHC Aerovox. We have loads of these in stock, but when people compare, all want to use Nippon Chemicon instaed.

Maybe because most of the larger NCC caps are made by Sprague in USA

 

[ghemink]

ghemik: I think they have a very clear, but still relaxed sound to them. Compared with other 'standard grade' brands, i find them to be the best i have tried so far. And IMO better than also some socalled 'high grade' brands like BHC Aerovox. We have loads of these in stock, but when people compare, all want to use Nippon Chemicon instaed.

Maybe because most of the larger NCC caps are made by Sprague in USA

Hi Lars,

Thanks for your feedback. I'll keep buying Nippon Chemicons with a much better feeling

And as I plan to use a lot of them in parallel, using a virtual 4-pole RCR type of setup, they should perform pretty well and filter remaining HF from the SMPS very well (in fact they do). The 4-pole setup would also prevent HF from the amp power supply rails to goback into the SMPS.

Just a question, would it make sense to make a virtual 4-pole for the GND connection of each cap as well, so having the GND of the cap connected to power supply GND via a small R and then go again with small R to the GND that goes to the amp. This would also "isolate" the power supply GND and amp GND in a similar way as real 4-pole caps would do. I plan to use at least 6 caps via RCR in parallel with R is 0.1Ohm.

Any ideas/advice

best regards

Gertjan

 

[Lars Clausen]

gertjan: I have never tried that, from a technical point of view it's not necessary, but then you never know, it might work well

 

[ghemink]

gertjan: I have never tried that, from a technical point of view it's not necessary, but then you never know, it might work well

Hi Lars,

Decided not to do it this time, the PCB that I have for the supply caps is not big enough to have all the caps and all those resistors on board. Will limit it to RCR in the rails as you have on your virtual 4-pole supply.

Best regards

Gertjan

 

[Zvon]

ZAPPulse

Hi Lars,

Some time ago I tried 2.2SE modules with electrostatic speakers with quite impressive results. The modules outperformed a reputable class A amplifier convincingly.

Would you recommend (any) modification of the Zap's when used for driving ESL's considering their highly capacitive & reactive nature?

To narrow down the load conditions let's assume Quad ESL63's (988 and 989 are quite similar). Their impedance, phase and equivalent RLC circuits are readily available on WWW.

Regards,
Zvon

 

[Lars Clausen]

Zvon: The ZAPpulse modules specialize in driving highly inductive or capacitive loads with no problems.

You should not have to modify the modules at all.

I can only recommend using short leads between the modules and power supply, to minimize the sonic influence from inherent power supply pumping, found generally on half bridge Class D amplifiers.

All the best from Denmark

 

[Zvon]

Thanks Lars,

The module photo is attached.

The power supply leads will be 3mm dia. copper wire (filed on one side to fit the ZAP mounting holes).
The wire length will not exceed 25mm (1").
Double oridnary bridges (35A/400V) with 22nf across each diode.
Power supply capacitors are 22,000uF/63V RIFA's, will bridge them with a smaller high capacity films or ELCO's.


Regrards,
Zvon

 

[Lars Clausen]

Zvon: Super Cool layout

 

[OzBradW]

Hi Lars

I was hoping you could help, i have been having some issues with your amp modules, my fault not the design i am sure.

I have been having some issues with ground loops, and have fried a 2.3se module while trying to eliminate the loop. The large 2.2 ohm wire wound resistor the metal oxide varistor have melted.

Can you advise if replacing these will resurrect the module or is it scrap now.

If it is worth an attempt to replace them could you please advise the component values of them. inparticular the Watt rating of the wire wound resistor and the brand of the Varistor used so i can source the same or equivalent

Thanks

Brad

 

[Lars Clausen]

Brad: I should think you can bring alive your module by changing the resistor and VDR.

The fault is owing to self oscillations from incorrect ground connections.

You can see a correct wiring plan for the zappulse amplifier here:

http://www.lcaudio.dk/zpactionpackwiring.pdf

The resistor should be a 3 watt flame proof type with low inductance. The original is a welwyn series 84, you can buy at Farnell.

The VDR is a 60 Volts original type is Epcos, but others will do nicely too. If you are unsure of the make of VDR, better use a 80 V type instead.

 

[OzBradW]

Thanks for the promt response Lars.

I will let you know if i was able to resuscitate the module.

Brad

 

[OzBradW]

Lars

I have sourced the replacement components and will be fitting them over the weekend.

One other question if i could, when testing the module, i have previously had a load (speaker) connected up. Do i also have to have the inputs either connected, unconnected or shorted to avoid damage to the unit?

Thanks for the link to for the wiring connections, I am trying something different to this (use in a subwoofer with eq circuits) and have been reconsidering my connections of all these circuits to the amp to remove the ground loop hum.

Brad

 

[Lars Clausen]

OzBradW: If assembled according to the plans, its not necessary to connect the inputs or outputs. In other cases it is better to connect inputs and outputs to minimize the risk of (unintended) oscillation. This goes for any kind of wideband power amplifier
(> 50 kHz).

 

[richie00boy]

Lars,

I am currently developing a high end sub-woofer for a small UK company. I would like to use the ZAPpulse because of it's compact size, ease of mounting and high power output. Have you considered a 'subwoofer' version of the module with lower Fc filter so that even less residual is present?

Also, please can you explain the difference between 2.3 and 2.3SE. I know 2.2 and 2.2SE used different caps and the SE tolerated higher voltage hence could put out more power. I could not find the information on your website regarding the newer versions, however.

 

[Lars Clausen]

Richieboy: The same rules apply with the 2.3 as the 2.2, with these regards.

I will consider the subwoofer version, i don't know if it's beneficial or not to go down that path. You will not have any side effects when using a ZAPpulse with residuals, for a subwoofer, or a full range system for that matter.

Best regards

Lars Clausen

 

[OzBradW]

Lars

Thanks, the module lives.

I now have a spare module as i bought a replacement thinking that it was totalled.

Hmmm now to find a use.

I have managed to eliminate the ground loop hum, but i still have a slight hum from the speakers (3 x 12" Shivas, see post under loudspeakers, subwoofers, 3 x 12" Shiva Sub) that i beleive is due to the topology of my setup and also the fact that i have low power circuits (EQ Graphic and EQ boost) close to the high powered circuits in an unsheilded enclosure. I have a new plan to fix that though (I hope, more tweeking).

All in all, for my current use i am extremely happy with the performance of your module, and its durability when treated as badly as i did.

Thanks again.

Brad

 

[using_e]

Separate Gate Drive supply

Does the separate gate drive supply only save a few watts thermal loss or does it also lead to some improvement in audio quality? And if so would it give additional improvement if the gate drive supply would be regulated?
Any comment especially from Lars are welcome.

using_e