FuriousD,
Hereby a FFT of 7.2kHz at 20Vrms in 8 Ohm load, of the UcD180.
I am sorry I can't give info about OEM customers they will not appreciate this. But when the amps are ready we can show you some details about the product.
Gertjan,
The difference between direct measurement of the UcD modulator and with a NE5532 is none. The first measurements showed with the NE5532 it was even better, but this was a measurement direclty after switched. So measurement with or without the capacitors, I do not get any difference in measurements.
As far as I know there are a lot of things to hear buy we can't measure. Becuase of this a lot of tech guys don't believe in listening to electronic components. I was one them, but I have now to admit that there are more things what you can hear but can not measure. Perhaps somebody can jump in and give some other theoratical info.
Reagrds,
Jan-Peter
www.hypex.nl
Hereby a FFT of 7.2kHz at 20Vrms in 8 Ohm load, of the UcD180.
I am sorry I can't give info about OEM customers they will not appreciate this. But when the amps are ready we can show you some details about the product.
Gertjan,
The difference between direct measurement of the UcD modulator and with a NE5532 is none. The first measurements showed with the NE5532 it was even better, but this was a measurement direclty after switched. So measurement with or without the capacitors, I do not get any difference in measurements.
As far as I know there are a lot of things to hear buy we can't measure. Becuase of this a lot of tech guys don't believe in listening to electronic components. I was one them, but I have now to admit that there are more things what you can hear but can not measure. Perhaps somebody can jump in and give some other theoratical info.
Reagrds,
Jan-Peter
www.hypex.nl
Attachments
Jan-Peter said:
Hi Jan-Peter,
I`m also an electrical engineer and in general I used not to believe in things that we can not measure but that we can only hear, are we really hearing the difference or are we thinking that we hear a difference because we should since we changed something. However, I have to admit, many things you would expect to be not significant from an engineering point of view can make audible differences. I do not have sophisticated test equipment at home, so I can not measure that much anyway. Still I think measurements are very valuable. If we hear something that we can`t measure then I think the problem is in the measurements method. We should then find the right parameter to measure.
Anyway, enough philosophy for now. I`ll just try it sometime. Just good to know that the FFT results do not show any differences.
Best regards
Gertjan
xman737 said:Hi ghemink,
Thanks! Looking forward to your pictures.
Hi xman737,
Here are two pictures taken from two different angles. The whole setup is still very experimental as you can see and this is sure not the final setup.
The first picture shows the view from the power supply side. I used a 250W transformer that gives about 2x32V AC output voltage. I ordered it at partsexpress in the US, the transformer should give 2x35V but in Japan the line voltage is only 100V while it is 115V in the US so I get a few volts less. The small PCB at the right of the transformer is a softstart module (from LCaudio), this module puts a series resistor in between the power supply and the transformer during start up and shorts the resistor after a few seconds. As a result, the peak currents for charging the power supply caps is reduced. Then just behind the transformer is the PCB with the diode bridge rectifiers and power supply caps of 10.000uF each. This gives me about 42V DC after rectification. This module is also from LCaudio, it is the Predator power supply that was designed to be used with the LCaudio ZAPpulse modules. It works perfect for the UCD180 as well. In the back are two UCD180 modules.
In the other picture (in the next post), you can see the two UCD180 modules from the other side. As you can see, I have soldered all wires directly to the connectors or to the PCB (power supply). I did this because I did not have the connectors and I prefer soldering above having connectors. I realize that I may not be able to return the modules in case of any trouble with them, however, I took that risk. I have used WBT RCA connectors and speaker binding posts, but I`ll be using XLR connectors in the final version. In the final version I`ll probably add a few more 10.000uF capacitors in the power supply.
Best regards
Gertjan
Attachments
xman737 said:
Hi xman,
You are welcome. Building the active system takes time, progressing slowly. The speakers are almost ready (not yet in child-proof version so I set them up only at night in the living room), amps will take still a bit more time but I can listen to the speakers with my Marantz amp (4 channel Tripath) so I`m not without music all the time
. They sound better than my passive B&W CDM9NT so I don`t really want to listen to those speakers anymore since my half finished 3-way really sounds quite a bit better. Good luck with your project.
Best regards
Gertjan
Hi all!
Jan-Peter, about 7.2k..
my experiments had the floating balance between 2nd and 3nd harmonics, which was changing by the changing loop components, however THD was kept (almost). Would be the even/odd balance different from sample to sample of UcD180 due to the tolerance worst case, with still around 0.013% THD?
PS:..Looking forward FuriousD analyse for your plot...
Jan-Peter, about 7.2k..
my experiments had the floating balance between 2nd and 3nd harmonics, which was changing by the changing loop components, however THD was kept (almost). Would be the even/odd balance different from sample to sample of UcD180 due to the tolerance worst case, with still around 0.013% THD?
PS:..Looking forward FuriousD analyse for your plot...
Ivan,
We never tested a batch of the UcD180 with this FFT analyze. It's indeed a good idee to test a batch.
What do you mean with;
Regards,
Jan-Peter
www.hypex.nl
We never tested a batch of the UcD180 with this FFT analyze. It's indeed a good idee to test a batch.
What do you mean with;
Regards,
Jan-Peter
www.hypex.nl
Hi ghemink
You're setup looks great. That soft start module from LC looked like a nice buy until I saw the price, it's the same cost as the UCD180?? Looks like it's worth about 5E at the most. The predator power supply I had a difficult time finding information on. I think they removed it from their site, (to replace it with something even more expensive) but I saw a reference on it that seemed to be going for something like 300E..
What do you get for that, 2 caps, 2 full bridge rectifiers.. I hope they built it on a gold slate.
Am I wrong or are they a huge rip off? I've seen others on here using their products, I assumed it was for the value (decent price) but apparently not, so what's the attraction?
You're setup looks great. That soft start module from LC looked like a nice buy until I saw the price, it's the same cost as the UCD180?? Looks like it's worth about 5E at the most. The predator power supply I had a difficult time finding information on. I think they removed it from their site, (to replace it with something even more expensive) but I saw a reference on it that seemed to be going for something like 300E..
What do you get for that, 2 caps, 2 full bridge rectifiers.. I hope they built it on a gold slate.
Am I wrong or are they a huge rip off? I've seen others on here using their products, I assumed it was for the value (decent price) but apparently not, so what's the attraction?
classd4sure said:
Hi classd4sure,
Just go to http://www.lcaudio.com/ scroll down the page and you should see the price info of the modules. The power supply costs 53.65 Euro (they have 0 shipping costs).
The reason I bought them is that I did not have the time to go to Akihabara to buy all the parts myself and build something on an experimental board. I preferred the modules to avoid a lot off work and time. I think for 53.65 Euro they are still not cheap but in my case I considered it OK. I ordered them together with the ZAPpulse modules that are also far more expensive than UCD but they have the potential for large output power when bridged and driving 4 Ohm. Their output coil is oversized. I plan to use the ZAPpulse for the dipole woofers and bridge two of them and drive two 8 Ohm woofers in parallel (so 4 Ohm bridged). That is a lot of power that the UCD can not deliver (not yet, will be in the future).
Yes, the softstart module is a bit expensive as well, I agree. In the long past when I was a student, I fabricated stuff like that myself. However, at the moment I do not have any capability anymore to make PCBs myself so I basically have not much choice. I don want to start again fiddling around with FeCl3, UV exposure stuff etc to make my own PCBs.
Best regards
Gertjan
Hi Ghemink,
Whew~ Power supply is "only" 53.65 euros. That's a bit better isn't it.
I fully sympathise not wanting to make the pcb, I'm in the same boat, and it's often worth paying a bit extra for some added convenience. This is why I"ll have no problem paying for a UCD module instead of trying to make my own as it would cost me more to do so also.
Even still, that soft start circuit is a tempting little device, I wanted to buy one, seems it's 40euros and not the 60 I thought, still way overpriced but .. not as bad as 60. I wonder what Jan-Peter could make one for? 10 amps isn't enough anyway, I'll make my own on some perf board..
Their power supply I was mistakingly looking at was some "virtual four pole" that included soft start and dc filter for 315euros.
It's little wonder they can afford to ship worldwide for free at those prices.
Regards,
Chris
Whew~ Power supply is "only" 53.65 euros. That's a bit better isn't it.
I fully sympathise not wanting to make the pcb, I'm in the same boat, and it's often worth paying a bit extra for some added convenience. This is why I"ll have no problem paying for a UCD module instead of trying to make my own as it would cost me more to do so also.
Even still, that soft start circuit is a tempting little device, I wanted to buy one, seems it's 40euros and not the 60 I thought, still way overpriced but .. not as bad as 60. I wonder what Jan-Peter could make one for? 10 amps isn't enough anyway, I'll make my own on some perf board..
Their power supply I was mistakingly looking at was some "virtual four pole" that included soft start and dc filter for 315euros.
It's little wonder they can afford to ship worldwide for free at those prices.
Regards,
Chris
classd4sure said:Hi Ghemink,
Whew~ Power supply is "only" 53.65 euros. That's a bit better isn't it.
I fully sympathise not wanting to make the pcb, I'm in the same boat, and it's often worth paying a bit extra for some added convenience. This is why I"ll have no problem paying for a UCD module instead of trying to make my own as it would cost me more to do so also.
Even still, that soft start circuit is a tempting little device, I wanted to buy one, seems it's 40euros and not the 60 I thought, still way overpriced but .. not as bad as 60. I wonder what Jan-Peter could make one for? 10 amps isn't enough anyway, I'll make my own on some perf board..
Their power supply I was mistakingly looking at was some "virtual four pole" that included soft start and dc filter for 315euros.
It's little wonder they can afford to ship worldwide for free at those prices.
Regards,
Chris
Hi Chris,
The 10amp for the softstart is limited by the relay. Probably it is possible to replace the relay with one that can handle more current. The bottom side of that PCB (you can`t see that in the picture) also has a number of SMD components that also cost a few cents
Gertjan
power supply
this will be an easy question to answer, im sure.
how does one determine the power supply requirements?
if I was going to run a pair of UcD180 modules, would a power supply consisting of an appropriately rated toroid, rectifiers and smoothing caps be sufficient?
and how is the size of the toroid determined? ill be running into 8 ohm speakers. im assuming both modules can runn from the same supply.
i have a spare 330VA 45-0-45 toroid, and was wondering if it was up to the job. it'd save me some money if i could use it...
cheers,
ben
this will be an easy question to answer, im sure.
how does one determine the power supply requirements?
if I was going to run a pair of UcD180 modules, would a power supply consisting of an appropriately rated toroid, rectifiers and smoothing caps be sufficient?
and how is the size of the toroid determined? ill be running into 8 ohm speakers. im assuming both modules can runn from the same supply.
i have a spare 330VA 45-0-45 toroid, and was wondering if it was up to the job. it'd save me some money if i could use it...
cheers,
ben
Re: power supply
Hi Ben,
The Toroid that you are having giving 45-0-45 Volt would give a too high voltage after rectification (about 62V or so). That would be too much for the UCD180. The maximum power supply voltage for the UCD180 is specified at + and - 50V. So a transformer that gives 35-0-35 would be close to the edge, giving about 48V DC after rectification. I personally would take a bit more margin and go for say a 30-0-30 transformer, that would give about 41V power supply. If you can get for example 32-0-32 or something like that, that would be perfect. The capacitors on the UCD180 that are connected to the power rails are all rated at 50V, so I would not go higher than that, better stay a bit lower.
I use two UCD180 modules with a 250W toroid and 2x10.000uF, that works fine. I`m still in the process of building the amps, everything is still experimental. I also plan to modify the modules, replacing most of the caps with black gates.
These mods will take time, will go on a business trip from tomorrow and I also want to listen to the modules for sometime without any mods. Will then do the mods and `see` if I can hear the `before/after` difference.
Best regards
Gertjan
anomalous said:
Hi Ben,
The Toroid that you are having giving 45-0-45 Volt would give a too high voltage after rectification (about 62V or so). That would be too much for the UCD180. The maximum power supply voltage for the UCD180 is specified at + and - 50V. So a transformer that gives 35-0-35 would be close to the edge, giving about 48V DC after rectification. I personally would take a bit more margin and go for say a 30-0-30 transformer, that would give about 41V power supply. If you can get for example 32-0-32 or something like that, that would be perfect. The capacitors on the UCD180 that are connected to the power rails are all rated at 50V, so I would not go higher than that, better stay a bit lower.
I use two UCD180 modules with a 250W toroid and 2x10.000uF, that works fine. I`m still in the process of building the amps, everything is still experimental. I also plan to modify the modules, replacing most of the caps with black gates.
These mods will take time, will go on a business trip from tomorrow and I also want to listen to the modules for sometime without any mods. Will then do the mods and `see` if I can hear the `before/after` difference.
Best regards
Gertjan
Bruno Putzeys said:
Hi Bruno,
Thanks for the additional info. What about the caps that are connected to the power rails. Can they handle that 50V (or maybe have to handle upto 55V due to net voltage variations). I don`t know how much margin is designed in by the manufacturer of the caps.
Best regards
Gertjan
Hmm, I don't happen to have one handy so I don't know what caps are used on it. If they say 50V it's indeed a good idea to stay below this value. In that case, you should count that with no signal and 10% mains over-voltage, 50V is the limit. Which is more or less what you suggested earlier 45V at nominal mains.
45V at nominal mains.
Remember that operating 120V MOSFET’s at 110V (+/- 55V) is running very close to the edge – only leaving +/- 5V for Ringing, PSU pumping and line regulation.
IR recommends a 35% headroom – I believe this is too much, but anything less then 10% I believe is too little….
D/S over-voltage even for nS’s (Ringing) well breakdown a FET. Although manufactures design a small “headroom” into their designs to increase production yields, this should not be relied on.
As Bruno recommends, +/- 50V should really be the limit.
John
IR recommends a 35% headroom – I believe this is too much, but anything less then 10% I believe is too little….
D/S over-voltage even for nS’s (Ringing) well breakdown a FET. Although manufactures design a small “headroom” into their designs to increase production yields, this should not be relied on.
As Bruno recommends, +/- 50V should really be the limit.
John