mikeks said:
let's have them old chap....
Okay. Attached is the simulation file for the 200 W class d amp with pfc front end and rail balance circuit ... (please change the extension from .txt back to .asc to make it run in LTspice)
By the way, LTspice works great and is a
free download from http://www.linear-tech.com/software/Modded UcD180
Hello all,
I have modified two of my UcD180 modules and built two otherwise identical stereoamps. One with standard UcD180 module and one with the modified ones.
The mods are not exactly the same as posted earlier.
I have changed the 22uF caps in the voltage regulators to 22uF BG (as previous), have changed the 22uF coupling caps between opamps and UcD module to 4.7uF BG-N (as previous). Have added two 22uF caps BG after the voltage regulators, from + rail to GND and - rail to GND. Have also added 2 x 0.1uF BG-N directly at the power pins of the opamp and to GND. I have removed the BG caps that I previously had included as extra power supply decoupling for the power transistor stage.
So how does this sound. Well, I spent hours listening swtiching back and forth. I can tell you that the difference is very/very small. First of all, the 4.7uF coupling caps are a little bit too smal since the input impedance of the UcD on the - input can become about 1kOhm when driven symmetrically. So this gives a roll of point at about 34Hz at the - input. So this will result in slightly less bass output below 34Hz, but not that much since it will only drop 6dB max becuase the + input will start to fall off at a much lower frequency (about 4Hz since the impedance at + input of the UcD is 10k). This could be the reason why I had the feeling that the original unmodified module sounded sometimes a bit warmer.
With music that did not have much low frequency content, this would not play a role. Then I had the feeling that the modified version had slightly improved resolution and a bit more accurate imaging at the highest frequencies. However, I became less sure of that after 2-3 hours or so. I think I got tired of plugging in and out those loudspeaker cables. Anyway, the difference if any is very small. And without any doubt, far less than the differences between my other amps that are immediately audible.
The conclusion does not change, the UcD180, delivered as is, is a very good sounding power amp.
It also stays very cool. I had run those amps for a couple of hours at relatively high listening levels in a room that was pretty hot (31C) since I did not want to have noise from the airco during listening. The temperature of the modules was only about 41C (measured with a small thermocouple), so only 10C higher than the ambient temperature. The modules were mounted on aluminum plate of 25x25cm and about 2mm thick. I switched between speakers with the amps on (of course no input signal), could not hear any clicks, so DC output must be low. I measured it, about 5-15mV for 3 modules and about 50mV for the 4th module, very good I would say.
Anyway, very happy with these modules.
Best regards
Gertjan
Hello all,
I have modified two of my UcD180 modules and built two otherwise identical stereoamps. One with standard UcD180 module and one with the modified ones.
The mods are not exactly the same as posted earlier.
I have changed the 22uF caps in the voltage regulators to 22uF BG (as previous), have changed the 22uF coupling caps between opamps and UcD module to 4.7uF BG-N (as previous). Have added two 22uF caps BG after the voltage regulators, from + rail to GND and - rail to GND. Have also added 2 x 0.1uF BG-N directly at the power pins of the opamp and to GND. I have removed the BG caps that I previously had included as extra power supply decoupling for the power transistor stage.
So how does this sound. Well, I spent hours listening swtiching back and forth. I can tell you that the difference is very/very small. First of all, the 4.7uF coupling caps are a little bit too smal since the input impedance of the UcD on the - input can become about 1kOhm when driven symmetrically. So this gives a roll of point at about 34Hz at the - input. So this will result in slightly less bass output below 34Hz, but not that much since it will only drop 6dB max becuase the + input will start to fall off at a much lower frequency (about 4Hz since the impedance at + input of the UcD is 10k). This could be the reason why I had the feeling that the original unmodified module sounded sometimes a bit warmer.
With music that did not have much low frequency content, this would not play a role. Then I had the feeling that the modified version had slightly improved resolution and a bit more accurate imaging at the highest frequencies. However, I became less sure of that after 2-3 hours or so. I think I got tired of plugging in and out those loudspeaker cables. Anyway, the difference if any is very small. And without any doubt, far less than the differences between my other amps that are immediately audible.
The conclusion does not change, the UcD180, delivered as is, is a very good sounding power amp.
It also stays very cool. I had run those amps for a couple of hours at relatively high listening levels in a room that was pretty hot (31C) since I did not want to have noise from the airco during listening. The temperature of the modules was only about 41C (measured with a small thermocouple), so only 10C higher than the ambient temperature. The modules were mounted on aluminum plate of 25x25cm and about 2mm thick. I switched between speakers with the amps on (of course no input signal), could not hear any clicks, so DC output must be low. I measured it, about 5-15mV for 3 modules and about 50mV for the 4th module, very good I would say.
Anyway, very happy with these modules.
Best regards
Gertjan
Gertjan,
Nice test between an unmodified UcD180 and a modified....
Indeed the modules stays quit cool (ICE Cool Blue Power), the most heat will be in the outputchoke and the voltage regulators. The heat of the choke can reach +/-57oC. Because of this we advise to have some airflow in the metalcase.
Samsagaz,
The UcD400 will be soon available, please check regular the website of the manufacture.
Cheers,
Jan-Peter
www.hypex.nl
Nice test between an unmodified UcD180 and a modified....
Indeed the modules stays quit cool (ICE Cool Blue Power
), the most heat will be in the outputchoke and the voltage regulators. The heat of the choke can reach +/-57oC. Because of this we advise to have some airflow in the metalcase.Samsagaz,
The UcD400 will be soon available, please check regular the website of the manufacture
.Cheers,
Jan-Peter
www.hypex.nl
Jan-Peter said:Gertjan,
Nice test between an unmodified UcD180 and a modified....
Indeed the modules stays quit cool (ICE Cool Blue Power
Samsagaz,
The UcD400 will be soon available, please check regular the website of the manufacture
Cheers,
Jan-Peter
www.hypex.nl
Hi Jan-Peter,
Next time I`ll also measure the temperature of the output choke.
Next experiment will be to drive the UcD directly from the preamp, so bypassing the opamps. My preamp has less than 1mV DC out, so should be OK (measured it). The supplier of the preamp also said the output uses DC servo`s, no coupling caps. If this does not produce an audible difference, then it does not make sense to me to experiment with further tweaking of the input opamps.
Best regards
Gertjan
ghemink said:
Hello all,
Today I did the experiment where I bypassed the opamps and coupling caps and connected the XLR input signal directly to the UcD module.
The effect is audible but it is small. The sound is a bit more open and a bit more transparant.
The module that had a DC output voltage of about 50mV two days ago, now got a DC output voltage of about 120mV when both the + and - input of the UcD module were shorted. With open input, the DC output of that module was about 70mV. The other module had only 5mV DC output. My preamp did not add DC output. I was a bit afraid of power supply pumping because of that 120mV DC offset. So I was carefull. First only connected 39Ohm resistors to the output and monitored the power supply voltage and DC offset. That was OK, no pumping. Then I connected my 4 Ohm speakers, also OK, no pumping at all. Then I did the listening tests. I cranked up the signal from the preamp 13dB to compensate for the gain loss.
That`s all. I guess the AD8620 will have a similar effect judging from the mail from Alain sometime ago.
Best regards
Gertjan
Finally I have mine playing (with OPA627 pre on top):
Just finished I think I can still tell what is coming. The amp is far better under stress than anything I've had earlier. Even at loud levels without being compressed or loosing control. I knew the speaker (HFLS-1, Swedish forum project with Vifa DX25 and PL18) could perform well but they had still been somewhat dull and boring through my Denon AVR-1803 which at higher volume was really compressed and lost a lot of the dynamics so that the music just colapsed to a goo without life or sparkle. Any notch up in volume made all instrument sound one step closer to eachother. This is hardly noticable with the UcD180 amp and the dynamic sound and resolution follows me all the way up until the wallpaper starts turning into dust. The low end is really tight with perfect control and does not lose steam like what the Denon suffered with.
Technically? Both the Denon and the UcD amp should be fairly equivalent in terms of maximum power so the comparison is fair I think. The modules look and feel very genuine and build quality is great. The power-on feature is permanently connected and switching on produces only the slightest little chirp or pop from the speakers. The DC offset was 10 mV for the left and 35 mV for the right channel. With my ear up close to the speakers there is not the faintest buzz or crackle. I'm surprised as I was expecting at least a little whizzing but nothing. I don't mind the silence...
The sound quality so far is excellent. The Bösendorf sounds great. Tower of Power is making my leg stomp. The Boston Pops are in my living room. I'll see what happens with time but I've already decided to build another one with UcD180 and one with UcD400.
The next ones will use better parts for the supply as the modules by all means are worth and deserve it.
An externally hosted image should be here but it was not working when we last tested it.
Just finished I think I can still tell what is coming. The amp is far better under stress than anything I've had earlier. Even at loud levels without being compressed or loosing control. I knew the speaker (HFLS-1, Swedish forum project with Vifa DX25 and PL18) could perform well but they had still been somewhat dull and boring through my Denon AVR-1803 which at higher volume was really compressed and lost a lot of the dynamics so that the music just colapsed to a goo without life or sparkle. Any notch up in volume made all instrument sound one step closer to eachother. This is hardly noticable with the UcD180 amp and the dynamic sound and resolution follows me all the way up until the wallpaper starts turning into dust. The low end is really tight with perfect control and does not lose steam like what the Denon suffered with.
Technically? Both the Denon and the UcD amp should be fairly equivalent in terms of maximum power so the comparison is fair I think. The modules look and feel very genuine and build quality is great. The power-on feature is permanently connected and switching on produces only the slightest little chirp or pop from the speakers. The DC offset was 10 mV for the left and 35 mV for the right channel. With my ear up close to the speakers there is not the faintest buzz or crackle. I'm surprised as I was expecting at least a little whizzing but nothing. I don't mind the silence...
The sound quality so far is excellent. The Bösendorf sounds great. Tower of Power is making my leg stomp. The Boston Pops are in my living room. I'll see what happens with time but I've already decided to build another one with UcD180 and one with UcD400.
The next ones will use better parts for the supply as the modules by all means are worth and deserve it.
Hi !
you are talking about high volumes, what about extremely low volumes ? do you steel feel deep bass and understandable lyrics ? If you make this test beetween a single ended class A and an class AB you will feel the differnce... Does a great class D wins ?
and you can even measure it if you have a multimeter to see how many volts you needs on the same loudspeaker to be "understandable"
you are talking about high volumes, what about extremely low volumes ? do you steel feel deep bass and understandable lyrics ? If you make this test beetween a single ended class A and an class AB you will feel the differnce... Does a great class D wins ?
and you can even measure it if you have a multimeter to see how many volts you needs on the same loudspeaker to be "understandable"
sound quality at various levels
PA,
My amp based on UcD180 modules preserves transparency and detail even at low-moderate levels (which is my normal listening level). Bass is still well defined even at low levels, but of course at low levels energy in the middle of the frequency range becomes perceptually more prominent, due to the way the human hearing system deals with loudness. None of the amps I tried before in my home (class AB amps, plus another so-called class "T" amp) achieved this. Never tried a class A amp though.
Goliardo
PA,
My amp based on UcD180 modules preserves transparency and detail even at low-moderate levels (which is my normal listening level). Bass is still well defined even at low levels, but of course at low levels energy in the middle of the frequency range becomes perceptually more prominent, due to the way the human hearing system deals with loudness. None of the amps I tried before in my home (class AB amps, plus another so-called class "T" amp) achieved this. Never tried a class A amp though.
Goliardo
Yes... under lock and key...samsagaz said:
There is a thread called 'Development of a "reference" class D starting point' in which several posters concentrated on piecing together a working circuit based on what's known from the UcD patent description, sometimes gently nudged by the circuit's designer.
The schematics/layout data of finished UcD modules like the ones that are commercially sold won't be given away anytime soon though. Their development has required a lot of investment in time and money and it would be economically unsound to give it all away just like that.
I do think the discussion in the "Reference" thread should largely satisfy your need for understanding the basic operation of the circuit. Diligent experimentation should actually get you a working amp. If I'm not mistaken, one or two regular posters (and I don't mean myself and Jan-Peter) have already pulled it off.
ZAPpulse versus UcD
I know many people are interested in this. So finally I did it. I built up a stereo ZAPpulse based amp using exactly the same power supply as I use for my UcD180 amp.
I have listened and switched back and forth between the UcD and the ZAPpulse amp. The UcD was a slightly modded version (see earlier description) where I have replaced some caps with BG caps, sorry, I don`t have an unmodded UcD anymore and it would be too much work to restore it to the original state which would of course be a more fair comparison. The ZAPpulse was a 2.2SE version, but the older one with the 47uF BG caps, the newer version uses 100uF Vishay caps.
My listening tests were limited to only 1-2hr or so. This is probably too short to have a good judgement and there was a lot of switching back and forth between the two amps. I don`t really like this way of testing. I prefer to listen for a few weeks to one amp and then listen to another amp. This way, one learns how a certain amp sounds with different CDs so that in my opinion, small differences can be better diagnosed. Both amps were connected via XLR with the same type of cable (ALphacore Micropurl silver) to my DEQX PDC that I use as a preamp for this test.
After listening, I have to conclude that both amps sound very good. I can not say that I have a preference for one or the other based on listening only. I could not find a significant difference in sound quality. Now of course my hearing has limitations and I`m not used to real high-end equipment and my speakers are also not that high-end (B&W CDM9NT) so other people maybe able to find differences between them and on different speakers there maybe differences between them.
There are however some differences that are not that much related to the sound quality directly:
1. The UcD180 has cleaner switching off behavior. During the tests, I did not use the "standby" feature that both amps have. So both amps were switched on and off by switching on and off the power. During switching on, both amps behave quite similarly, they switch on with a very small click sound, probably because of some DC at the output. This is not disturbing. Then switching off was different, the UcD switches off very cleanly, a small click or so, but that was it. The ZAPpulse switches of with some chirping sounds (not very loud, so not damaging) coming out of the speakers, this is caused by interference between the two ZAPpulse modules, this maybe reduced by improving the GND connections between the two modules, but I did not do that tonight, however, in the past I have checked that and it reduced the effect. When the amp is switched on, one can not hear any interference like effects, so it is only a switching off issue. The ZAPpulse also gives some low frequency output signal during switching off, I see my woofers moving a bit when switching off, maybe this is because the negative power supply draws more current and therefore the negative supply voltage collapses faster? And even more funny was that during switching off, the ZAPpulse picked up some signal from a radio or TV station, I could hear that for a few seconds (at low level but audible at listening position). This was repeatable throughout the session. Probably this goes away when the amps are built in in a metal case. Nothing of this was audible when the amps were in the switched on-state. So if one finds this annoying, it is probably best to switch the amp on and off using the "standby" function (I have not tried that). In fact, when I put my ears almost in the tweeters, I can hear a bit of chirping with the UcD180 as well, of course also only during switching off, but it is at a very/very low level, so suggesting that interference like effects with the UcD modules are less
2. The noise level of the UcD180 is lower than that of the ZAPpulse. The ZAPpulse has a similar noise level as my Tripath based Marantz and also similar gain as the Tripath. The gain of the UcD180 is 7dB lower but to my ears the noise level of the UcD180 is much lower than that 7dB. In fact, with the UcD, the noise is limited by the preamp noise, when I siwtch off the preamp it is almost impossible to hear the noise of the amp, I have to put my ear in the tweeter to be able to hear a trace of noise, very/very low noise level. With the ZAPpulse, the noise is ZAPpulse limited. Probably one of the reasons is that the UcD itself has a gain that is about 20dB lower than the ZAPpulse (giving less noise), the UcD module therefore uses two opamps to crank up the input signal that goed into the UcD circuit.
Another difference is the components used in the ZAPpulse and the UcD. The ZAPpulse has more components and has for example a DCservo circuit that maybe handy for use with certain speakers. The components on the ZAPPulse seem to be of high quality with BG caps at critical places. The ZAPpulse PCB also seems to be of high quality, many times soldering power supply wires on and off the print did not harm the soldering islands on the PCB. Probably based on the choice and number of components, the price differnce maybe justified. A pre of the ZAPpulse is that it can deliver enormous output power (at least on paper, have not checked it), and it can be bridged to get more than 1000W with a 4 Ohm load, this maybe a pre for some users. On the other hand, the UcD180 looks like a simple circuit, relatively few components and no things like a DC servo circuit. PCB looks good and has connectors for the power and signal lines, I prefer soldering however and so I soldered everything.
In my case, I have to use the amps for an active 3-way system, for mid and high I prefer the UcD modules mainly because of the lower noise level. The switching off behavior of the UcD is also better, but that can probably be made good with the ZAPpulse as well by using the "standby" option.
For woofers, however, the ZAPpulse maybe preferred when very high power levels are needed in a bridge configuration.
Best regards
Gertjan
I know many people are interested in this. So finally I did it. I built up a stereo ZAPpulse based amp using exactly the same power supply as I use for my UcD180 amp.
I have listened and switched back and forth between the UcD and the ZAPpulse amp. The UcD was a slightly modded version (see earlier description) where I have replaced some caps with BG caps, sorry, I don`t have an unmodded UcD anymore and it would be too much work to restore it to the original state which would of course be a more fair comparison. The ZAPpulse was a 2.2SE version, but the older one with the 47uF BG caps, the newer version uses 100uF Vishay caps.
My listening tests were limited to only 1-2hr or so. This is probably too short to have a good judgement and there was a lot of switching back and forth between the two amps. I don`t really like this way of testing. I prefer to listen for a few weeks to one amp and then listen to another amp. This way, one learns how a certain amp sounds with different CDs so that in my opinion, small differences can be better diagnosed. Both amps were connected via XLR with the same type of cable (ALphacore Micropurl silver) to my DEQX PDC that I use as a preamp for this test.
After listening, I have to conclude that both amps sound very good. I can not say that I have a preference for one or the other based on listening only. I could not find a significant difference in sound quality. Now of course my hearing has limitations and I`m not used to real high-end equipment and my speakers are also not that high-end (B&W CDM9NT) so other people maybe able to find differences between them and on different speakers there maybe differences between them.
There are however some differences that are not that much related to the sound quality directly:
1. The UcD180 has cleaner switching off behavior. During the tests, I did not use the "standby" feature that both amps have. So both amps were switched on and off by switching on and off the power. During switching on, both amps behave quite similarly, they switch on with a very small click sound, probably because of some DC at the output. This is not disturbing. Then switching off was different, the UcD switches off very cleanly, a small click or so, but that was it. The ZAPpulse switches of with some chirping sounds (not very loud, so not damaging) coming out of the speakers, this is caused by interference between the two ZAPpulse modules, this maybe reduced by improving the GND connections between the two modules, but I did not do that tonight, however, in the past I have checked that and it reduced the effect. When the amp is switched on, one can not hear any interference like effects, so it is only a switching off issue. The ZAPpulse also gives some low frequency output signal during switching off, I see my woofers moving a bit when switching off, maybe this is because the negative power supply draws more current and therefore the negative supply voltage collapses faster? And even more funny was that during switching off, the ZAPpulse picked up some signal from a radio or TV station, I could hear that for a few seconds (at low level but audible at listening position). This was repeatable throughout the session. Probably this goes away when the amps are built in in a metal case. Nothing of this was audible when the amps were in the switched on-state. So if one finds this annoying, it is probably best to switch the amp on and off using the "standby" function (I have not tried that). In fact, when I put my ears almost in the tweeters, I can hear a bit of chirping with the UcD180 as well, of course also only during switching off, but it is at a very/very low level, so suggesting that interference like effects with the UcD modules are less
2. The noise level of the UcD180 is lower than that of the ZAPpulse. The ZAPpulse has a similar noise level as my Tripath based Marantz and also similar gain as the Tripath. The gain of the UcD180 is 7dB lower but to my ears the noise level of the UcD180 is much lower than that 7dB. In fact, with the UcD, the noise is limited by the preamp noise, when I siwtch off the preamp it is almost impossible to hear the noise of the amp, I have to put my ear in the tweeter to be able to hear a trace of noise, very/very low noise level. With the ZAPpulse, the noise is ZAPpulse limited. Probably one of the reasons is that the UcD itself has a gain that is about 20dB lower than the ZAPpulse (giving less noise), the UcD module therefore uses two opamps to crank up the input signal that goed into the UcD circuit.
Another difference is the components used in the ZAPpulse and the UcD. The ZAPpulse has more components and has for example a DCservo circuit that maybe handy for use with certain speakers. The components on the ZAPPulse seem to be of high quality with BG caps at critical places. The ZAPpulse PCB also seems to be of high quality, many times soldering power supply wires on and off the print did not harm the soldering islands on the PCB. Probably based on the choice and number of components, the price differnce maybe justified. A pre of the ZAPpulse is that it can deliver enormous output power (at least on paper, have not checked it), and it can be bridged to get more than 1000W with a 4 Ohm load, this maybe a pre for some users. On the other hand, the UcD180 looks like a simple circuit, relatively few components and no things like a DC servo circuit. PCB looks good and has connectors for the power and signal lines, I prefer soldering however and so I soldered everything.
In my case, I have to use the amps for an active 3-way system, for mid and high I prefer the UcD modules mainly because of the lower noise level. The switching off behavior of the UcD is also better, but that can probably be made good with the ZAPpulse as well by using the "standby" option.
For woofers, however, the ZAPpulse maybe preferred when very high power levels are needed in a bridge configuration.
Best regards
Gertjan
Hi Gertjan,
there is one thing I have noticed when comparing amps with the DEQX PDC as preamp. I don't know if it is that critical when using it as fullrange amp but if you want to compare them in a three-way setup be sure to make two measurements under the same conditions, one with each amp in place. Then make two exactly the same correctionfilters and load them into the PDC. This is because the PDC also compensates for any gain, frequency and phase response differences in the amplifiers.
Do you think there will be any interference between the amp modules when building a six channel amp ? The Zappulse provides an external clock input for that to keep them in sync.
Does anyone have compared this modules (or the Zappulse) with the better Classe A or single ended triode amps out there ? (excuse if this question has been asked before, the thread is becoming to big to read it all) That's what I am interested in (well personally, I allready have SE triode amps). I have never heard an AV receiver (even some very expensive ones) sounding good so that's very easy to beat with DIY. In my very humble opinion I find all of the Denon's Yamaha's, Marantzes etc sounding crap, there are a lot of DIY projects out there wich sounds much better than that.
there is one thing I have noticed when comparing amps with the DEQX PDC as preamp. I don't know if it is that critical when using it as fullrange amp but if you want to compare them in a three-way setup be sure to make two measurements under the same conditions, one with each amp in place. Then make two exactly the same correctionfilters and load them into the PDC. This is because the PDC also compensates for any gain, frequency and phase response differences in the amplifiers.
Do you think there will be any interference between the amp modules when building a six channel amp ? The Zappulse provides an external clock input for that to keep them in sync.
Does anyone have compared this modules (or the Zappulse) with the better Classe A or single ended triode amps out there ? (excuse if this question has been asked before, the thread is becoming to big to read it all) That's what I am interested in (well personally, I allready have SE triode amps). I have never heard an AV receiver (even some very expensive ones) sounding good so that's very easy to beat with DIY. In my very humble opinion I find all of the Denon's Yamaha's, Marantzes etc sounding crap, there are a lot of DIY projects out there wich sounds much better than that.
Details are outlined elsewhere in this thread. In short, you do indeed ground the - input but you can enjoy the added advantage of the differential input if you ground the - input *at the preamplifier's ground*. Use a shielded twisted pair cable to do this. It is unnecessary to get a balanced driver.
On the power amp side, the two wires go to +input and -input, the shield goes to gnd. On the preamp side, the -wire and shield both go to ground and the +wire obviously goes to the signal.
Most of us fit the power amps with a female XLR connector and make an adapter cable with an RCA connector on one side and a male XLR on the other side. On the RCA side wire A=signal, wire B and shield=GND. On the XLR side pin1=shield (gnd), pin2=wire A, pin3=wire B.
On the power amp side, the two wires go to +input and -input, the shield goes to gnd. On the preamp side, the -wire and shield both go to ground and the +wire obviously goes to the signal.
Most of us fit the power amps with a female XLR connector and make an adapter cable with an RCA connector on one side and a male XLR on the other side. On the RCA side wire A=signal, wire B and shield=GND. On the XLR side pin1=shield (gnd), pin2=wire A, pin3=wire B.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.