Hi Ronny.
That looks interesting. Though I have a UCD180HG V2 board which looks different.
Gain (Voltage?) of 4.5 might do at 0db=1.6V =7.2V ends up at 3.2W at my 16R/98db/W speakers.
Better choice would be to use my Zapfilter in differential mode. I'll see later how it works.
Anyhow - I need to figure out how to do the tweak on my modules.
Step one: I would like to leave the coupling caps in and just bypass the LM4562.
BTW: I have the funny feeling that it can't be RG=R10 on my module. R10 is sitting on the Regs.R11/R12 (should be R12/R13 I guess ) seems also to be wrong There must be an error in the Hypex application note.
Hi There,
Rg on the UCD 180 HG V-2 module is R14 (Rf=R12,R13). I think you need around 800R for gain of 10.
Regards,
Peter.
Hi Peter.
In fact you are right. I've been in contact with Hypex today.
I am thinking to use a gain of 4.5 (which is 13db). I wouldn't need the input stage from an amplification perspective.
However, Hypex told me instead of passing the opamp by I should rather run a gain of 0 otherwise I'd get the impedance down to 1k8.
Now I've got a couple of more questions:
1. To make the input stage gain 0 , can I just take out RG=R14=inf?
2. What do I have to consider in case of SE operation if IN- goes to ground and RG is open?
THX
In fact you are right. I've been in contact with Hypex today.
I am thinking to use a gain of 4.5 (which is 13db). I wouldn't need the input stage from an amplification perspective.
However, Hypex told me instead of passing the opamp by I should rather run a gain of 0 otherwise I'd get the impedance down to 1k8.
Now I've got a couple of more questions:
1. To make the input stage gain 0 , can I just take out RG=R14=inf?
2. What do I have to consider in case of SE operation if IN- goes to ground and RG is open?
THX
Yep - knew the formula. Guess what I meant with R14=inf=infinite
I was just wondering if it can be that easy.
I wanted to make sure that I don't mess anything up in SE operation.BTW. Do I really need a heatsink even if I just draw not more then a couple of Watts?
Yep - knew the formula. Guess what I meant with R14=inf=infinite
I was just wondering if it can be that easy.
BTW. Do I really need a heatsink even if I just draw not more then a couple of Watts?
Hi Soundcheck,
I haven't checked it, but I can't see why it would affect SE operation. You just decrease gain of the buffer stage. BTW Rg = 0, not inf. You just need to short it in place of original R14.
Peter
Koifarm,
You should never ever wire an input like that, unless you want to hear how good your cables don't sound. You're using them as a low pass filter, no doubt it sounds smooth, and truncated.
You're also risking the infamous pin 1 problem. This really is not the way.
I don't think it's PSRR you're risking, though the op amp stage likely boosts it quite a bit as it has some 120db PSRR plus the regulators. The modulator has 60dB PSRR flat across the audioband which is actually very decent. Remember it's not 80dB at 100Hz that plumets to 10dB by 10kHz.
What you really ruin is CMRR, and you also bypass the anti aliasing /RF filters that every good input stage should have. Using your cable as a low pass filter is not a substitute for this.
Really it's a terrible example as it has no regard to signal integrity whatsoever and blindly relies on the less is more theory which is just wrong.
Regards,
John
You should never ever wire an input like that, unless you want to hear how good your cables don't sound. You're using them as a low pass filter, no doubt it sounds smooth, and truncated.
You're also risking the infamous pin 1 problem. This really is not the way.
I don't think it's PSRR you're risking, though the op amp stage likely boosts it quite a bit as it has some 120db PSRR plus the regulators. The modulator has 60dB PSRR flat across the audioband which is actually very decent. Remember it's not 80dB at 100Hz that plumets to 10dB by 10kHz.
What you really ruin is CMRR, and you also bypass the anti aliasing /RF filters that every good input stage should have. Using your cable as a low pass filter is not a substitute for this.
Really it's a terrible example as it has no regard to signal integrity whatsoever and blindly relies on the less is more theory which is just wrong.
Regards,
John
Dear john,
i am not a man of figures but of experience. I use Hypex modules for more than 9 years( build more than 20 times ) without the opamp stage. And my experience is that there is less noise and i like the sound. Coupling the module like my way is no problem, i do not need 2 Capacitors in a row. My preamp has a big output Cap and the hypex has also a big input cap. So i have bypassed the hypex one. About driving 1k8 of impedance you just have to try if you preamp can do that( or check it first ). But having no noise in my tweeters ( 101 db ) with just leaving some parts is the best. Further i do not have to buy the expensive modules who are build with the better regulators to lower the noise, i can just buy the ST versions and tweak some capacitors.
" less is more " is no theory, it is an experience.
There are no theories who can describe how something sounds you just have to listen. I know this can be a shock for people how buy or build things just for the great spec's. But the best things i heard had not always the best specs. I say good specs is a good base but listen and make your own decision is the most important thing.
By the way the opampless Hypex with and without tube buffer is used worldwide by privates and professionals in home and studio. And this is the best reference you can get.
Ronny
i am not a man of figures but of experience. I use Hypex modules for more than 9 years( build more than 20 times ) without the opamp stage. And my experience is that there is less noise and i like the sound. Coupling the module like my way is no problem, i do not need 2 Capacitors in a row. My preamp has a big output Cap and the hypex has also a big input cap. So i have bypassed the hypex one. About driving 1k8 of impedance you just have to try if you preamp can do that( or check it first ). But having no noise in my tweeters ( 101 db ) with just leaving some parts is the best. Further i do not have to buy the expensive modules who are build with the better regulators to lower the noise, i can just buy the ST versions and tweak some capacitors.
" less is more " is no theory, it is an experience.
There are no theories who can describe how something sounds you just have to listen. I know this can be a shock for people how buy or build things just for the great spec's. But the best things i heard had not always the best specs. I say good specs is a good base but listen and make your own decision is the most important thing.
By the way the opampless Hypex with and without tube buffer is used worldwide by privates and professionals in home and studio. And this is the best reference you can get.
Ronny
Last edited:
koifarm,
I prefer facts over fallacy and I do not let someone's experience or subjective perception stand in the way of them. As such I'd very much appreciate you refrain from stacking fibs and fallacies one upon the other to make your point. For example you haven't been using Hypex modules for "more than 9 years", they haven't even existed that long. Check the birthdate of this thread for quick evidence of that.
Also building 20 or 20 million amps the exact same, wrong way, for your "friends", doesn't grant you more expertise than did the first one. If you're also to make a habit of such wild statements as privates and professionals using your method worldwide, to grant either it or yourself credibility, I will have to demand evidence of your polling, and the names to go with it.
I'm telling you why there can be less "perceived" noise and audio to go with it, and you like tubes anyway, so after coupling them to an otherwise neutral and transparent amp you should like the sound. Guido Tent sells tube-ucd hybrid kits for these modules, and I'll bet you anything he doesn't wire them like you do. Oh, if only there was some way we could find that out.
http://www.tentlabs.com/Products/DI...d_amp_mounting_manual_V1.0 5-9-2009_small.pdf
Notice he doesn't compromise his modules as you've done? That's because he knows better.
"Less is more" is simply an opinion and taken out of proper context it will usually be wrong as it especially is in this instance. You're not going to convince me you improved signal integrity or sound by circumventing the RF/anti alias filtering, or by cutting the legs off the common mode noise rejection ability of the module. Interfacing is a complete science and it's not up for subjective preferential debate. The module is designed in such a way that allows the very best methods that you'd be wise to make use of.
Your point about how you can buy the cheapest edition and cheaply tweak it up to your preference level is interesting, but I think only from a profit margin perspective, not for achieving the best possible results. That you find no improvement with the superior module should be indicative of an issue with your methods, and it seems clear you lack the expertise to make the most of them. Furthermore it is likely you didn't find improvement with using your methods on the high grade modules because they've already far surpassed them.
Bypassing the coupling caps alone, particularly on the standard versions where they seem to use electrolytic coupling caps versus the film caps of the higher quality version, explains your subjective improvement by itself.
Interestingly enough you needn't shoot the legs off your modules in order to improve on the coupling means.
Johnny
I prefer facts over fallacy and I do not let someone's experience or subjective perception stand in the way of them. As such I'd very much appreciate you refrain from stacking fibs and fallacies one upon the other to make your point. For example you haven't been using Hypex modules for "more than 9 years", they haven't even existed that long. Check the birthdate of this thread for quick evidence of that.
Also building 20 or 20 million amps the exact same, wrong way, for your "friends", doesn't grant you more expertise than did the first one. If you're also to make a habit of such wild statements as privates and professionals using your method worldwide, to grant either it or yourself credibility, I will have to demand evidence of your polling, and the names to go with it.
I'm telling you why there can be less "perceived" noise and audio to go with it, and you like tubes anyway, so after coupling them to an otherwise neutral and transparent amp you should like the sound. Guido Tent sells tube-ucd hybrid kits for these modules, and I'll bet you anything he doesn't wire them like you do. Oh, if only there was some way we could find that out.
http://www.tentlabs.com/Products/DI...d_amp_mounting_manual_V1.0 5-9-2009_small.pdf
Notice he doesn't compromise his modules as you've done? That's because he knows better.
"Less is more" is simply an opinion and taken out of proper context it will usually be wrong as it especially is in this instance. You're not going to convince me you improved signal integrity or sound by circumventing the RF/anti alias filtering, or by cutting the legs off the common mode noise rejection ability of the module. Interfacing is a complete science and it's not up for subjective preferential debate. The module is designed in such a way that allows the very best methods that you'd be wise to make use of.
Your point about how you can buy the cheapest edition and cheaply tweak it up to your preference level is interesting, but I think only from a profit margin perspective, not for achieving the best possible results. That you find no improvement with the superior module should be indicative of an issue with your methods, and it seems clear you lack the expertise to make the most of them. Furthermore it is likely you didn't find improvement with using your methods on the high grade modules because they've already far surpassed them.
Bypassing the coupling caps alone, particularly on the standard versions where they seem to use electrolytic coupling caps versus the film caps of the higher quality version, explains your subjective improvement by itself.
Interestingly enough you needn't shoot the legs off your modules in order to improve on the coupling means.
Johnny
Johnny, first of all you need to read better.
I said i use Hypex modules not Hypex class D modules for more than 9 years. And second i never said i build a better UCD but i said i build a UCD i like the most. After using ST and HG modules for a view years i did not like them anymore. After modifying the UCD with tube buffers i discovered a new sound with i like much more.
Once again it is all about the sound, you need to listen and then decide for yourself what you like. You do not need anybody to tell you what to buy/build. And sharing experience that is why we use this forum.
These ones are the benefits:
- You do not need HG version, just the ST version. Because i do not use the opamp stage i do not need the expensive HXR regulators.( it is much cheaper)
- I have just one (de)coupling condensator in the audio path.
- Because i need 4,5( instead of 1) Vrms to drive the UCD, cables pickup less noise.
Just one fact hypex build the OEM versions of his ucd modules without the opamp stage attached. So that oem builders can use their own stages.
On the pictures i use Hypex OEM modules these you can not buy as private person.
Ronny
An externally hosted image should be here but it was not working when we last tested it.
I said i use Hypex modules not Hypex class D modules for more than 9 years. And second i never said i build a better UCD but i said i build a UCD i like the most. After using ST and HG modules for a view years i did not like them anymore. After modifying the UCD with tube buffers i discovered a new sound with i like much more.
Once again it is all about the sound, you need to listen and then decide for yourself what you like. You do not need anybody to tell you what to buy/build. And sharing experience that is why we use this forum.
An externally hosted image should be here but it was not working when we last tested it.
These ones are the benefits:
- You do not need HG version, just the ST version. Because i do not use the opamp stage i do not need the expensive HXR regulators.( it is much cheaper)
- I have just one (de)coupling condensator in the audio path.
- Because i need 4,5( instead of 1) Vrms to drive the UCD, cables pickup less noise.
Just one fact hypex build the OEM versions of his ucd modules without the opamp stage attached. So that oem builders can use their own stages.
An externally hosted image should be here but it was not working when we last tested it.
On the pictures i use Hypex OEM modules these you can not buy as private person.
Ronny
Last edited:
Ronny,
I fear you've truly an aversion to the facts. Clearly this discussion pertains to ucd. Going back years prior to that is, to be generous, irrelevant, non sequitur. It also doesn't leave you with any less of a common impedance concern, or any other anomaly one is likely to experience by wiring the shield directly to signal reference, which as a rule, you never do! Hypex has an app note which demonstrates appropriate methods, you should read it, this is not a question of preference.
As a manufacturer using OEM modules I find it unlikely you'd be concerned with the savings of not having the HXR regulators, since OEM modules don't come with them anyway. Also please note my comments in no way reflect upon your preference for a tube hybrid approach, but rather the method used to employ it, which from your first picture that you keep referring others to, can be greatly improved.
Your facts once again fall short of reality. OEM versions do include an input stage, along with the additional means to bypass them. They may not easily be available for purchase, but you don't need any special credentials to download the datasheet and see that. A true OEM should be skilled enough to put that feature to use without greatly compromising it. For example the edge connector allows for a certain configuration that wouldn't compromise RF performance even when using an offboard input stage. Certainly in their wisdom they'd also include any necessities that bypassing the onboard input stage may have curtailed, at least one would hope.
Soundcheck, I think the proper way to make the input stage a stable unity gain buffer is to short the feedback resistors, and remove the gain setting resistor, leaving it open. The other alternative discussed may not have the kind of results you're after.
Johnny
I fear you've truly an aversion to the facts. Clearly this discussion pertains to ucd. Going back years prior to that is, to be generous, irrelevant, non sequitur. It also doesn't leave you with any less of a common impedance concern, or any other anomaly one is likely to experience by wiring the shield directly to signal reference, which as a rule, you never do! Hypex has an app note which demonstrates appropriate methods, you should read it, this is not a question of preference.
As a manufacturer using OEM modules I find it unlikely you'd be concerned with the savings of not having the HXR regulators, since OEM modules don't come with them anyway. Also please note my comments in no way reflect upon your preference for a tube hybrid approach, but rather the method used to employ it, which from your first picture that you keep referring others to, can be greatly improved.
Your facts once again fall short of reality. OEM versions do include an input stage, along with the additional means to bypass them. They may not easily be available for purchase, but you don't need any special credentials to download the datasheet and see that. A true OEM should be skilled enough to put that feature to use without greatly compromising it. For example the edge connector allows for a certain configuration that wouldn't compromise RF performance even when using an offboard input stage. Certainly in their wisdom they'd also include any necessities that bypassing the onboard input stage may have curtailed, at least one would hope.
Soundcheck, I think the proper way to make the input stage a stable unity gain buffer is to short the feedback resistors, and remove the gain setting resistor, leaving it open. The other alternative discussed may not have the kind of results you're after.
Johnny
Johnny, i just can tell you try and listen. And publish here your experience.
Others can learn from that experience.
I used a long time the "ST" modules. And i like them after modifying as good as the "HG" modules. With the tubebuffer they were perfect for me. My tubebuffer can drive impedance as low as 600 Ohm but is optimzed for 1k8 (lowest distortion).
I know al the application papers of hypex. But there are more ways to Rome.
Before you know with way you like most you have to try them all.
I tried for example 8 different tubebuffers with more than 5 types tubes, but one of the simplest tubebuffers i like the most.
If you happy to do it your way, that is great.
I do it my way.
Ronny
Others can learn from that experience.
I used a long time the "ST" modules. And i like them after modifying as good as the "HG" modules. With the tubebuffer they were perfect for me. My tubebuffer can drive impedance as low as 600 Ohm but is optimzed for 1k8 (lowest distortion).
I know al the application papers of hypex. But there are more ways to Rome.
Before you know with way you like most you have to try them all.
I tried for example 8 different tubebuffers with more than 5 types tubes, but one of the simplest tubebuffers i like the most.
If you happy to do it your way, that is great.
I do it my way.
Ronny
Hi folks.
I think Ronny is more then right on one point.
There is no opamp, which is not degrading the sound. So, I strongly believe
that if the source can cope with 1k8 amp input impedance and an overal gain of 4.5 is OK, this is the way to go.
I mean - If possible I would generally avoid any active element on the input. Tubes might have a nice sound signature, but are they lossless? No way!
The open question to me is still how to pass the input by on the UCD180HG module (by keeping it somwhat revertable) . What would be the best access point?
Cheers
I think Ronny is more then right on one point.
There is no opamp, which is not degrading the sound. So, I strongly believe
that if the source can cope with 1k8 amp input impedance and an overal gain of 4.5 is OK, this is the way to go.
I mean - If possible I would generally avoid any active element on the input. Tubes might have a nice sound signature, but are they lossless? No way!
The open question to me is still how to pass the input by on the UCD180HG module (by keeping it somwhat revertable) . What would be the best access point?
Cheers
Last edited:
Must there be a LED burning when the ucd180HG is stiwched on?
D6 is that a LED? on the ucd180hg print?
I have very low level sound, i put my ear to the speaker. But the volume will not go up.
What do i wrong, i Put the wires as in the drawing from hypes.
Also i connect the ON wires to GND,
is the UCD180HG broken?
thanks for any help
gr,
hans.
D6 is that a LED? on the ucd180hg print?
I have very low level sound, i put my ear to the speaker. But the volume will not go up.
What do i wrong, i Put the wires as in the drawing from hypes.
Also i connect the ON wires to GND,
is the UCD180HG broken?
thanks for any help
gr,
hans.
Hello,
I'm going to use my pair of UCD180 (standard) for a friend's wedding and would like to have recommendations regarding their use in that kind of environment : playing loud for a long time.
Though Hypex claim a "public adress application", they are only 100 watts (in 8ohms) and that's not so much even with 95db bass speakers.
Any input (other that "don't do that, you fool")?
I'm going to use my pair of UCD180 (standard) for a friend's wedding and would like to have recommendations regarding their use in that kind of environment : playing loud for a long time.
Though Hypex claim a "public adress application", they are only 100 watts (in 8ohms) and that's not so much even with 95db bass speakers.
Any input (other that "don't do that, you fool")
?
bypassing buffer stage on UcD180ST - How to do?
Hello all Hypex'ers,
I'm the happy owner of a DIY power amp based on UcD180HG HxR modules with a regulated power supply. I'm driving this amp directly from my DIY DAC (Buffalo 32S) with attenuation in the digital domain. I'm very satisfied with this combo but my tweaker genes tells me that maybe I can still improve......
I listen to music at modest levels and the combined gain in the DAC and amp is way higher than I actually need. I know I have different options including:
A) reducing the gain of the DAC I/V stage
B) reducing the gain of the UcD buffer stage
C) bypassing the buffer-stage of the UcD
Doing it by methods A and B is pretty straightforward, but as method C will reduce the number of active components in my signal path and remove coupling caps entirely, I have decided to follow this route.....
For comparison I will not be tweaking the UcD's modules I already have, but will get a pair of ST modules for the purpose. The DAC I have features balanced output with low DC offset (<1mV) and low output impedance (milliohms to a few ohms) so I believe that they should drive the UcD's just fine without the buffer stage (1K8?). As I'm using a regulated supply I think that supply pumping will not be an issue.....
So to get to the question: how do I actually wire this bypass on the UcD180ST modules the best way?
I have been going over pages and pages but I have not found detailed descriptions of how this is done. Could somebody please point me in the right direction?
Ideally I would like to use the signal connections on the ST modules for easy "swap" comparison between the bypassed ST and my "reference" HG/HxR.
I don't mind if the modding of the ST's will be irreversible.
All kind of advice would be very welcome!
Thanks,
Nic
Hello all Hypex'ers,
I'm the happy owner of a DIY power amp based on UcD180HG HxR modules with a regulated power supply. I'm driving this amp directly from my DIY DAC (Buffalo 32S) with attenuation in the digital domain. I'm very satisfied with this combo but my tweaker genes tells me that maybe I can still improve......
I listen to music at modest levels and the combined gain in the DAC and amp is way higher than I actually need. I know I have different options including:
A) reducing the gain of the DAC I/V stage
B) reducing the gain of the UcD buffer stage
C) bypassing the buffer-stage of the UcD
Doing it by methods A and B is pretty straightforward, but as method C will reduce the number of active components in my signal path and remove coupling caps entirely, I have decided to follow this route.....
For comparison I will not be tweaking the UcD's modules I already have, but will get a pair of ST modules for the purpose. The DAC I have features balanced output with low DC offset (<1mV) and low output impedance (milliohms to a few ohms) so I believe that they should drive the UcD's just fine without the buffer stage (1K8?). As I'm using a regulated supply I think that supply pumping will not be an issue.....
So to get to the question: how do I actually wire this bypass on the UcD180ST modules the best way?
I have been going over pages and pages but I have not found detailed descriptions of how this is done. Could somebody please point me in the right direction?
Ideally I would like to use the signal connections on the ST modules for easy "swap" comparison between the bypassed ST and my "reference" HG/HxR.
I don't mind if the modding of the ST's will be irreversible.
All kind of advice would be very welcome!
Thanks,
Nic