Hello,
My first post here, but I am a synth and recording studio DIY guy with lots of experience and absolutely no understanding of what I am doing
I am planning to build two channels of Zap Pulses powered by the predator PSU.
1. I have access to 1000VA power transformers, but what voltage do you think I should go for. I don't really need a lot of power, but do you think it sound a lot better in a higher power set up? I could go for whatever, but I just want to hear some thoughts about pros and cons.
I am also a bit worried about working with high voltage... I am planning to use a separate 15V transformer for the Gate Drive.
2. How big does it have to be?
3. Can I just split the power after the IEC input connector to both transformers?
I will connect this amp to a Yamaha DSP-E800 pre amp that also has center and rear amplifiers as well as DTS/DD decoders. They go dirt cheap at ebay for the moment and the specs seems good with 70W/channel in 8ohm and 24,000F capacitance in the power supply. And it is only for the effect channels anyway... the zaps will power my setup for music
Anyway, the amp has a power output that is controlled of the standby button of the unit. I wouldn't really like to power my Zap Pulse from it, but I thought I could build a relay circuit inside my zap that is controlled from this output which controls the zap's standby feature. In other words I would put my zap pulses in standby mode from the standby button on my preamp and remote
4. In that case I suddenly have three power transformers in my chassi, one power, one gate driver and one for the relay circuit. Would this be a problem?
Thanks for any help!
Robert
My first post here, but I am a synth and recording studio DIY guy with lots of experience and absolutely no understanding of what I am doing
I am planning to build two channels of Zap Pulses powered by the predator PSU.
1. I have access to 1000VA power transformers, but what voltage do you think I should go for. I don't really need a lot of power, but do you think it sound a lot better in a higher power set up? I could go for whatever, but I just want to hear some thoughts about pros and cons.
I am also a bit worried about working with high voltage... I am planning to use a separate 15V transformer for the Gate Drive.
2. How big does it have to be?
3. Can I just split the power after the IEC input connector to both transformers?
I will connect this amp to a Yamaha DSP-E800 pre amp that also has center and rear amplifiers as well as DTS/DD decoders. They go dirt cheap at ebay for the moment
and the specs seems good with 70W/channel in 8ohm and 24,000F capacitance in the power supply. And it is only for the effect channels anyway... the zaps will power my setup for music Anyway, the amp has a power output that is controlled of the standby button of the unit. I wouldn't really like to power my Zap Pulse from it, but I thought I could build a relay circuit inside my zap that is controlled from this output which controls the zap's standby feature. In other words I would put my zap pulses in standby mode from the standby button on my preamp and remote
4. In that case I suddenly have three power transformers in my chassi, one power, one gate driver and one for the relay circuit. Would this be a problem?
Thanks for any help!
Robert
An externally hosted image should be here but it was not working when we last tested it.
Picture of my Zappulse not finished cabinet but playing.And yes Lars Clausen there is an Zapsolute under it(beautiful sounding amp).I´m using a 2*50V 1000VA trafo and you soldered bigger capacitors to me on the Predator PSU to handle the bigger voltage 2*15000 80V and it works just fine and the amp sound fantastic at its price. Unfortunally as you know (i mailed you about it)the right module got some shorting or something it just blinks very heavily.I have returned it to you,i have mailed about the problem when i shut the amp down it give the speakers a real "puff" that`s what happend with the modul.It also took one of my sub in the Martin Logan SL3 speaker to the grave.Is there any easy solution to this "puff" when im shutting down the amp,because the sub`s in my speakers are expensive?Some people have no problems at all with this fenomen i seem to be the only one that really get a "puff" in the speakers have i done something wrong,or is it the bigger trafo and capacitors that makes this"puff"?
/ Sven-Åke
Wicked Zapplifier
Hello everyone ..
well since this happens to be the post of zappulse I would like to share my current idea that i have been having lately
allthough 3-phase is not legal use in diy in denmark .. as far as i know ... here is the idea, with 3 phases using 3 trannies each 2x42V 500Vac supplying 6 sets of circuits of the neverconnected kind ( 3 positive and 3 negative ) with a total of 180.000 uF's , connected to 2 High-Q. El's with short leads to the ZapModules
All together a very stable supply in my mind .. given that it would now be a tiny 300Hz ripple without any noise from the mains *SSSS*
And Lars .. Thanks a great many times for sending the Zapfilter replacement so quick .. *S* it's been implemented with great care in the Pioneer and works Super ( connected according to the PCM1716 scheme ... but the filter caps connected to EXTL-R should be removed, no problem then )
still tweaking on :
Hello everyone ..
well since this happens to be the post of zappulse I would like to share my current idea that i have been having lately
allthough 3-phase is not legal use in diy in denmark .. as far as i know ... here is the idea, with 3 phases using 3 trannies each 2x42V 500Vac supplying 6 sets of circuits of the neverconnected kind ( 3 positive and 3 negative ) with a total of 180.000 uF's , connected to 2 High-Q. El's with short leads to the ZapModules
All together a very stable supply in my mind .. given that it would now be a tiny 300Hz ripple without any noise from the mains *SSSS*
And Lars .. Thanks a great many times for sending the Zapfilter replacement so quick .. *S* it's been implemented with great care in the Pioneer and works Super ( connected according to the PCM1716 scheme ... but the filter caps connected to EXTL-R should be removed, no problem then )
still tweaking on :
Zappulse 2.2 SE oscilating
Hellow Mr. Clausen
I need your help.
A friend of mine bought (through the Greek distributor) three ZapPulse 2.2SE modules (he needs four more).
He set up a monoblock with one of them in a half steel, half aluminum enclosure using a 250 W 2x50 V AC X-former (+/- 69V DC), a 30 A bridge and 2x 15000uF with very sort thick wiring throughout following the directions of the data sheet. He phoned me saying the amp had a bright sound, not much of power and it was becoming warm quite a lot (aluminum bloch attached to massive aluminum plate). I asked him not to operate it and bring to me for testing with the osciloscope. I was suspecting self oscilation.
When he brought it home, I connected a dummy 6 Ohm load and shorted the input.
The output with stand-by ON was quite acceptable (short bursts of 3MHz 1.5mVpp oscilations repeating with a frequency of 50 KHz, superimposed on +3 mV DC).
The output with stand-by OFF was a continuous 6.5Vpp 500KHz sinusoid (~1W RMS/6 Ohm).
The same picture when i connected a 50 Ohm signal generator to the input. The audio signal was simply modulating the 6.5Vpp 500KHz sinusoid.
The 2.2 Ohm Zobel resistor was becoming hot within 5 minutes (with no input signal). I tried different connections to ground, but no difference.
I asked he bring the second module. I connected a 100pF across the + and - input. Then i connected this module to:
A) my workbench regulated PSU (+/-24V DC).
B) a 2x24V AC (+/- 34V DC)100W x-former with a 10A bridge and 2x2200 uF
C) the initial 250 W 2x50 V AC x-former (+/- 69V DC), a 30 A bridge and 2x 15000uF
all ouside of any box, with very short leads. Same picture, only the oscilation amplitude followed the PSU voltage (2.1Vpp for the +/- 24V DC, 3.2Vpp for the +/- 34V DC, 6.5Vpp for the +/- 69V DC).
We havn't tested the third module, but chances are it will perform the same.
Please advise
Regards
George Papageorgiou
Hellow Mr. Clausen
I need your help.
A friend of mine bought (through the Greek distributor) three ZapPulse 2.2SE modules (he needs four more).
He set up a monoblock with one of them in a half steel, half aluminum enclosure using a 250 W 2x50 V AC X-former (+/- 69V DC), a 30 A bridge and 2x 15000uF with very sort thick wiring throughout following the directions of the data sheet. He phoned me saying the amp had a bright sound, not much of power and it was becoming warm quite a lot (aluminum bloch attached to massive aluminum plate). I asked him not to operate it and bring to me for testing with the osciloscope. I was suspecting self oscilation.
When he brought it home, I connected a dummy 6 Ohm load and shorted the input.
The output with stand-by ON was quite acceptable (short bursts of 3MHz 1.5mVpp oscilations repeating with a frequency of 50 KHz, superimposed on +3 mV DC).
The output with stand-by OFF was a continuous 6.5Vpp 500KHz sinusoid (~1W RMS/6 Ohm).
The same picture when i connected a 50 Ohm signal generator to the input. The audio signal was simply modulating the 6.5Vpp 500KHz sinusoid.
The 2.2 Ohm Zobel resistor was becoming hot within 5 minutes (with no input signal). I tried different connections to ground, but no difference.
I asked he bring the second module. I connected a 100pF across the + and - input. Then i connected this module to:
A) my workbench regulated PSU (+/-24V DC).
B) a 2x24V AC (+/- 34V DC)100W x-former with a 10A bridge and 2x2200 uF
C) the initial 250 W 2x50 V AC x-former (+/- 69V DC), a 30 A bridge and 2x 15000uF
all ouside of any box, with very short leads. Same picture, only the oscilation amplitude followed the PSU voltage (2.1Vpp for the +/- 24V DC, 3.2Vpp for the +/- 34V DC, 6.5Vpp for the +/- 69V DC).
We havn't tested the third module, but chances are it will perform the same.
Please advise
Regards
George Papageorgiou
Re: Zappulse 2.2 SE oscilating
Hello George,
That 500kHz signal that is modulated by the input signal is exactly what you should see. The ZAPpulse self-oscillates at that frequency, this is exactly as it should be. The 500kHz signal's amplitude also should follow exactly the power supply voltage since before the output LC filter, the signal would be a squarewave with an amplitude equal to the power supply voltage. The amplitude of that signal is quite large, this is because the output LC filter on the ZAPPulse has a high cut-off frequency in the order of 150kHz or so. Therefore, that 500kHz signal is not so very well suppressed. I have worked with the ZAPpulse a bit as well. That zobel network resistor becomes quite warm, this is because that 500kHz signal that is at the output will result in a current flowing in that resistor and will therefore heat up that resistor.
So bottomline: I think there is nothing abnormal with your modules, this is as axpected and I had similar experience with the ZAPpulse.
Best regards
Gertjan
gpapag said:
Hello George,
That 500kHz signal that is modulated by the input signal is exactly what you should see. The ZAPpulse self-oscillates at that frequency, this is exactly as it should be. The 500kHz signal's amplitude also should follow exactly the power supply voltage since before the output LC filter, the signal would be a squarewave with an amplitude equal to the power supply voltage. The amplitude of that signal is quite large, this is because the output LC filter on the ZAPPulse has a high cut-off frequency in the order of 150kHz or so. Therefore, that 500kHz signal is not so very well suppressed. I have worked with the ZAPpulse a bit as well. That zobel network resistor becomes quite warm, this is because that 500kHz signal that is at the output will result in a current flowing in that resistor and will therefore heat up that resistor.
So bottomline: I think there is nothing abnormal with your modules, this is as axpected and I had similar experience with the ZAPpulse.
Best regards
Gertjan
At first i wish a prosperous 2005 to all of you.
Gertjan thank you for your reply. It eased my worries a lot, at least you informed me that the modules behave "as expected".
Yesterday night i was thinking that i should give them a try. So today evening after the traditional New Years Eve family meeting , i went to the lab, i switched on the Zappulse
monoblock and an inverted LM 1875 amp, connected both to dummy loads, fed them through a cd player in repeat mode and i went for a double espresso with my wife. Two hours later i returned for a listening test. The Zappulse (at low power levels to match the LM) was more controlled, faster and more transparent than the LM amp. At higher levels it was performing great. Very good control and detail. The loudspeaker was a DIY sealed 70 lt B139/ sealed 7 lt B110/T33 replaced by a cheap 1.5" cone tweeter for this test. The temperature of the Zap did not raise a lot (ferrite ~42 d.C, Zobel resistor ~60 d.C, FETs~40, X-former ~45 d.Celsius.
So the module behaves OK acoustically. The problem (as i see it ) is that electrically it is "dirty": At the coil's input there is an almost rail to rail 500KHz square wave (130 Vpp for 138V supply). At the coil's output there is a 7.6 Vpp 500KHz sinusoidal wave.
Problem no 1. This sinusoidal signal feeds directly and continuosly the tweeter. 7.6 Vpp is 0.9WRMS in 8 Ohms, 1.8 W RMS in 4 Ohms. Not a good thing. I wonder what would happen to the sound if the output LC filter on the ZAPPulse had a lower cut-off frequency, say 50kHz. The tweeters would love it.
Problem no 2. It seems that a big part of the energy that the output coil has to dissipate is radiated around. I probed with my scope (20 MHZ only bandwith) the input socket; a 50mVpp 500KHZ square wave.On the upper and lower flats there was a decaying 100mVpp 5Mhz signal, resulting in a nasty signal of 250mVpp. This, inside the enclosure. Outside the enclosure at the input plug, there was a 180mVpp 5MHz signal left . Half a meter apart, at the output of my 50 Ohm signal source which was feeding the module, the 5MHz signal was 140mVpp. Some way of more efficient electromagnetic shielding is required.
Problem no 3. On the 15000 uF per rail PSU caps i probed an equally nasty 350mVpp 500KHZsquare wave /5MHz 200mVpp signal. Through the bridge at the x-former secondary there was a 450mVpp 500KHZ square wave/5MHz 50mV signal. It seems that a lot of experimentation with type/size of caps is required for trimming off this dirt. Also the temperature rize at the x-former's core manifests eddy currents in the core induced by this high freq. signal. Again good quality x-former is required (a C type will work also as a low pass filter toward the primary side).
Problem no 4. On the 2.2 Ohm Zobel resistor there is a 1.2Vpp voltage drop of the 500KHz sinusoidal signal. By Ohms law this is 0.08W RMS power dissipation. This resistor becoming quickly very hot (indicating that there are in addition a lot of harmonics passing through) and located between the two electrolytic caps, will probably dry them out sooner or later. A possible remedy is to have it raised higher, above the caps with ceramic spacers.
It seems that the "Nothing comes free" applies to the PWM amps as well. The laws of physics...
Regards
George
Gertjan thank you for your reply. It eased my worries a lot, at least you informed me that the modules behave "as expected".
Yesterday night i was thinking that i should give them a try. So today evening after the traditional New Years Eve family meeting , i went to the lab, i switched on the Zappulse
monoblock and an inverted LM 1875 amp, connected both to dummy loads, fed them through a cd player in repeat mode and i went for a double espresso with my wife. Two hours later i returned for a listening test. The Zappulse (at low power levels to match the LM) was more controlled, faster and more transparent than the LM amp. At higher levels it was performing great. Very good control and detail. The loudspeaker was a DIY sealed 70 lt B139/ sealed 7 lt B110/T33 replaced by a cheap 1.5" cone tweeter for this test. The temperature of the Zap did not raise a lot (ferrite ~42 d.C, Zobel resistor ~60 d.C, FETs~40, X-former ~45 d.Celsius.
So the module behaves OK acoustically. The problem (as i see it ) is that electrically it is "dirty": At the coil's input there is an almost rail to rail 500KHz square wave (130 Vpp for 138V supply). At the coil's output there is a 7.6 Vpp 500KHz sinusoidal wave.
Problem no 1. This sinusoidal signal feeds directly and continuosly the tweeter. 7.6 Vpp is 0.9WRMS in 8 Ohms, 1.8 W RMS in 4 Ohms. Not a good thing. I wonder what would happen to the sound if the output LC filter on the ZAPPulse had a lower cut-off frequency, say 50kHz. The tweeters would love it.
Problem no 2. It seems that a big part of the energy that the output coil has to dissipate is radiated around. I probed with my scope (20 MHZ only bandwith) the input socket; a 50mVpp 500KHZ square wave.On the upper and lower flats there was a decaying 100mVpp 5Mhz signal, resulting in a nasty signal of 250mVpp. This, inside the enclosure. Outside the enclosure at the input plug, there was a 180mVpp 5MHz signal left . Half a meter apart, at the output of my 50 Ohm signal source which was feeding the module, the 5MHz signal was 140mVpp. Some way of more efficient electromagnetic shielding is required.
Problem no 3. On the 15000 uF per rail PSU caps i probed an equally nasty 350mVpp 500KHZsquare wave /5MHz 200mVpp signal. Through the bridge at the x-former secondary there was a 450mVpp 500KHZ square wave/5MHz 50mV signal. It seems that a lot of experimentation with type/size of caps is required for trimming off this dirt. Also the temperature rize at the x-former's core manifests eddy currents in the core induced by this high freq. signal. Again good quality x-former is required (a C type will work also as a low pass filter toward the primary side).
Problem no 4. On the 2.2 Ohm Zobel resistor there is a 1.2Vpp voltage drop of the 500KHz sinusoidal signal. By Ohms law this is 0.08W RMS power dissipation. This resistor becoming quickly very hot (indicating that there are in addition a lot of harmonics passing through) and located between the two electrolytic caps, will probably dry them out sooner or later. A possible remedy is to have it raised higher, above the caps with ceramic spacers.
It seems that the "Nothing comes free" applies to the PWM amps as well. The laws of physics...
Regards
George
gpapag said:
Hello George,
You may want to try the Hypex UcD400 modules, I expect those to be a lot cleaner with respect to EMI and power supply noise. They have a LC filter that filters at a lower frequency, so less high frequencies at the output and they don't need a zobel network.
Sorry have to go, wife and kids are waiting
Gertjan
gpapag said:
Hi George,
Have done only a few measurements in the past on ZAPpulse modules. You have to be very carefull with the scope probes. The short GND cable (10cm or so) of the scope easily picks up noise (I guess from the electromagnetic field generated by the filter coil) so what you measure may not be the actual signal on a certain node. If you move that GND wire a bit, the signal strength changes. I have not measured my UCD180 modules yet (have to bring a scope from work) but I think the UcD180 coils have less EM stray field.
Best regards
Gertjan
Gertjan
You are right. In high frequency circuits, test leads can induce noise. I have been fooled again in the past. But now, after the initial measurements, i try to scope in the differential mode using the scope probes of both channels but having their ground leads not connected to circuit under test.
In this particular case (Zappulse) the noise seems to be not related to ground problems or test leads.
Today i tried a very good quality 2x38V AC/3.5A x-former with the module. The temperature of the x-former did not go higher than 35 d. Celcius.
richie00boy
Thank you for this very important information. So practically speaking , problem No 1 is non-existent. Or is this very optimistic?
Off topic: Today it was the third day i was unable to connect to diyaudio.com. What was it all about? I realised that i am a bit adicted to it.
Regards
George
You are right. In high frequency circuits, test leads can induce noise. I have been fooled again in the past. But now, after the initial measurements, i try to scope in the differential mode using the scope probes of both channels but having their ground leads not connected to circuit under test.
In this particular case (Zappulse) the noise seems to be not related to ground problems or test leads.
Today i tried a very good quality 2x38V AC/3.5A x-former with the module. The temperature of the x-former did not go higher than 35 d. Celcius.
richie00boy
Thank you for this very important information. So practically speaking , problem No 1 is non-existent. Or is this very optimistic?
Off topic: Today it was the third day i was unable to connect to diyaudio.com. What was it all about? I realised that i am a bit adicted to it.
Regards
George
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.