I recently acquired 2 of these little modules and have been playing with them to assess performance. They run off 2 SMPS400 power supplies.
The sound quality seems very good, although not as good as my linear amp. Bass is very good, midrange less so.
I have been looking on the web and found some threads where mods to the amps are discussed. Think of things like:
- replacing the opamp with something better
- replace supply caps
- bridging input caps
- biasing opamp to single ended class A with a CRD
- replacing output cap wit something better, etc. etc.
Most of these discussion are quite old (from before the HG- line was introduced) and I wonder if these mods are still relevant to the current UcD-line of products. It would seem that most of the improvements have already been implemented in the HGs.
It would be nice if we could put together a list of tweaks.
The sound quality seems very good, although not as good as my linear amp. Bass is very good, midrange less so.
I have been looking on the web and found some threads where mods to the amps are discussed. Think of things like:
- replacing the opamp with something better
- replace supply caps
- bridging input caps
- biasing opamp to single ended class A with a CRD
- replacing output cap wit something better, etc. etc.
Most of these discussion are quite old (from before the HG- line was introduced) and I wonder if these mods are still relevant to the current UcD-line of products. It would seem that most of the improvements have already been implemented in the HGs.
It would be nice if we could put together a list of tweaks.
I watch them a lot, but pretty expensive and couldn't read the caps, which are they? Is the soundquality more sterile then an a/b amp?
The sound is actually not sterile at all. Very dynamic and clear, perhaps not as laid back as your current amp. I find that I seem to appreciate them more and more over time.
I believe they are quite cheap for what they can do.
I wonder if the schematics are floating on the net. In some discussions, people are referring to the circuit as if they have more information.
bypassing input buffer
I too have 180HG (HxR) running on SMPS400. They are dry and accurate sounding but I too feel that they lack something -a bit too sterile and disengaged.
Have been thinking about DC coupling and bypassing the input buffer altogether but I can't find a description on how to do so on the internet. I have done it successfully with 180ST modules but the 180HG seems to have a different layout...
Would anybody know which caps to bypass? C1 and C2 seems obvious if I want to keep the input buffer but I am less sure about C31 and C32. Are they decoupling between OPA4562 and the UcD modulator or are they PS decoupling for the OPA4562?
Any advise would be highly appreciated 🙂
I too have 180HG (HxR) running on SMPS400. They are dry and accurate sounding but I too feel that they lack something -a bit too sterile and disengaged.
Have been thinking about DC coupling and bypassing the input buffer altogether but I can't find a description on how to do so on the internet. I have done it successfully with 180ST modules but the 180HG seems to have a different layout...
Would anybody know which caps to bypass? C1 and C2 seems obvious if I want to keep the input buffer but I am less sure about C31 and C32. Are they decoupling between OPA4562 and the UcD modulator or are they PS decoupling for the OPA4562?
Any advise would be highly appreciated 🙂
DC coupling
Couldn't resist bypassing C1 and C2 with some pieces of copper foil.
The result is quite staggering! The dryness is completely gone 🙂
For DC free sources, this is a no-brainer and has to be heard to be believed.
If my suspicion is correct that C31 and C32 are not additional AC coupling, C1 and C2 were the only caps in the signal path of my electronics.
Will probably still try to bypass the input opamp once I am confident that I know how to do it. If anybody out there knows where on the UcD180HG boards it's best to hook up the signal wires when bypassing the input opamp, it would be very helpful!
cheers,
Couldn't resist bypassing C1 and C2 with some pieces of copper foil.
The result is quite staggering! The dryness is completely gone 🙂
For DC free sources, this is a no-brainer and has to be heard to be believed.
If my suspicion is correct that C31 and C32 are not additional AC coupling, C1 and C2 were the only caps in the signal path of my electronics.
Will probably still try to bypass the input opamp once I am confident that I know how to do it. If anybody out there knows where on the UcD180HG boards it's best to hook up the signal wires when bypassing the input opamp, it would be very helpful!
cheers,
Hypex guard the design, not surprising really, so no info available. I spent three days working out the Flying Mole amps so successfully repair those and am currently working on the Hypex. They are so inexpensive really, it seems hardly worthwhile.
Be very careful twiddling the dead time pot, they will die!
Be very careful twiddling the dead time pot, they will die!
Short the 10uf coupling caps. You can pull the caps body off and
solder the leads together.
Remove the op amp and connect the input pins to the output pins.
This retains the RF LPF on the input.
Thanks JonSnell and deano.
I completely understand why Hypex don't want to have their design floating freely around.
I have already bypassed the 10uF coupling caps. I didn't remove them. Just bypassed them with copper strips. I might remove them entirely at some point.
Bypassing the LM4562 seems quite a bit more 'invasive' than what I first anticipated. On the ST modules, it was simple to remove two caps and connect the signal wires instead. Pretty easy to do, and to undo 🙂
Would you by any chance know which legs on the LM4562 that are inputs and outputs? From the LM4562 specs sheet I have output A as leg 1 and output B as leg 7, but it has two sets of non-inv/inv inputs (for A and for B).
After some more time with these modules as DC coupled, I must admit that bypassing the LM4562 should mostly be for reducing system gain. As is, the sound is quite spectacular 🙂
I completely understand why Hypex don't want to have their design floating freely around.
I have already bypassed the 10uF coupling caps. I didn't remove them. Just bypassed them with copper strips. I might remove them entirely at some point.
Bypassing the LM4562 seems quite a bit more 'invasive' than what I first anticipated. On the ST modules, it was simple to remove two caps and connect the signal wires instead. Pretty easy to do, and to undo 🙂
Would you by any chance know which legs on the LM4562 that are inputs and outputs? From the LM4562 specs sheet I have output A as leg 1 and output B as leg 7, but it has two sets of non-inv/inv inputs (for A and for B).
After some more time with these modules as DC coupled, I must admit that bypassing the LM4562 should mostly be for reducing system gain. As is, the sound is quite spectacular 🙂
Download the UcD180 OEM documentation from the Hypex website.
It shows the schematics for the input stage.
The 180HG (HxR) input stage differs in the op amp used and the lack of jumpers to bypass the input stage.
It shows the schematics for the input stage.
The 180HG (HxR) input stage differs in the op amp used and the lack of jumpers to bypass the input stage.
Last edited:
Thanks again!
Mkay, it seems that the inv inputs are for nfb which leaves non-inv input A and B to be connected with their corresponding outputs. If I read this correctly, that is. This then dictates: pin 3 to pin 1 and pin 5 to pin pin 7.
If I instead hook up the signal wires post buffer, I would skip some PCB path length and the 100p RF shorting caps. Since the UcD output filter is set to f3@35kHz anyway, this might not be such a bad tradeoff...
Mkay, it seems that the inv inputs are for nfb which leaves non-inv input A and B to be connected with their corresponding outputs. If I read this correctly, that is. This then dictates: pin 3 to pin 1 and pin 5 to pin pin 7.
If I instead hook up the signal wires post buffer, I would skip some PCB path length and the 100p RF shorting caps. Since the UcD output filter is set to f3@35kHz anyway, this might not be such a bad tradeoff...
Before you take the further actions as you intended I would suggest you to do the following:
I would be eager to know your impressions after that.
Baki
- Replace the LM4562 with LME 49722 and
- Replace the existing output filter capacitors with WIMA MKP2 2x0.33uF per channel. One cap should be placed underside the pcb and connected with the other at the stock position.
I would be eager to know your impressions after that.
Baki
Last edited:
Thanks a lot Baki!
Very good suggestions that I will definitely explore.
I take it that you have done these mods on HxR'ed HG modules yourself, right.
Trying out audio fuses will probably be my first step since it is so easy. Are you using the standard $5 or the $20 deluxe?
I have read quite a few accounts regarding the output filter caps and their importance in switching amps. It seems that there are many positive reports on doing this on nc400 too. Definitely something that should be tried too!
Regarding the opamp I was under the impression that the LM4562 is already up there among the best. Is the 49722 in your opinion just that much better? - I have also considered trying discrete op amps, but they wouldn't fit my build (slim build with internal height 48mm)
Bypassing the op amp would lower the gain (which is mostly a good thing in my case), but the penalty would be much lower input impedance causing the inductive reactance in my signal cables to infuse more top end roll off and phase shift. I did like the sound of non-buffered UcD STs, but something is telling me that 1.8kohm vs 100kohm makes a difference...
cheers!
Very good suggestions that I will definitely explore.
I take it that you have done these mods on HxR'ed HG modules yourself, right.
Trying out audio fuses will probably be my first step since it is so easy. Are you using the standard $5 or the $20 deluxe?
I have read quite a few accounts regarding the output filter caps and their importance in switching amps. It seems that there are many positive reports on doing this on nc400 too. Definitely something that should be tried too!
Regarding the opamp I was under the impression that the LM4562 is already up there among the best. Is the 49722 in your opinion just that much better? - I have also considered trying discrete op amps, but they wouldn't fit my build (slim build with internal height 48mm)
Bypassing the op amp would lower the gain (which is mostly a good thing in my case), but the penalty would be much lower input impedance causing the inductive reactance in my signal cables to infuse more top end roll off and phase shift. I did like the sound of non-buffered UcD STs, but something is telling me that 1.8kohm vs 100kohm makes a difference...
cheers!
Tried swapping in a $5 Audio Create fuse earlier this week and it seems to have burned in now. It wasn't love at first, but after a few days I really enjoy what it does: Clearer, cleaner, crisper and more detailed sound! IMHO it is 5 bucks VERY well spent and definitely worth to consider for other Hypex SMPS DIY'ers 🙂
Thanks again for the advice Baki.
cheers,
Thanks again for the advice Baki.
cheers,
Thanks Koifarm! I used your pictures as reference when I bypassed the opamps on my ST modules.
Yes, less than 8Vrms from the source won't squeeze the last juice out of the UcDs, but with efficient speakers I would very rarely need more than the 2Vrms that my Ciunas delivers.
I must admit that I am pleasantly surprised by how the SMPS makes the UcDs sound compared to the R-core and four pole cap linear I had in the chassis before. They are different in nature. The SMPS makes them sound more 'rigid' and 'linear' whereas the linear made them more forgiving giving a touch of warmth and umpf in the lower register and a different top end presentation with perhaps a tad more presence/weight while the SMPS makes for more extension. Its a tough call as I like both 🙂 The greatest asset of the linear approach is that the DIY'er has more options to tweak the resulting sound -which of course is a double-edged sword 😉
Koifarm, you use tube buffers as replacement for the onboard buffer, right? Have you compared that approach to the more recent UCD HG with HxR modules?
cheers,
Last edited:
Juhleren, good how you discribe the differrences of the power supplies.
It is always a mather of taste.
A weak point of the UCD ST and HG was the rejection of switchingnoise in the powersupply for the input opamp, therefore the HXR modules are a must when you like to use the input opamps.
Knowing this i have gone an other way. Bypass the input opamp or use oem boards. So i can use my own dedicated power supply for the input buffer. After testing several opamp with HXR options and tube buffers, i like the tube buffer most. This is my personal taste. Other options are technical better. But i can not listen longer than one hour to a unmodified UCD without getting irritated. In the beginning they sound spectaculer but after some time they are irritating me.
Now i use UCD for professional on stage amplification and for bassamplification in home, further i use as less digital equipment as possible.
A warning when modding the UCD's, they do not like DC on the input when the Caps are removed or bypassed( UCD output goes pumping). When using tubes you need a delay at the input of the UCD so that the tube first needs to warmup before connecting the UCd to the tubestage.
It is always a mather of taste.
A weak point of the UCD ST and HG was the rejection of switchingnoise in the powersupply for the input opamp, therefore the HXR modules are a must when you like to use the input opamps.
Knowing this i have gone an other way. Bypass the input opamp or use oem boards. So i can use my own dedicated power supply for the input buffer. After testing several opamp with HXR options and tube buffers, i like the tube buffer most. This is my personal taste. Other options are technical better. But i can not listen longer than one hour to a unmodified UCD without getting irritated. In the beginning they sound spectaculer but after some time they are irritating me.
Now i use UCD for professional on stage amplification and for bassamplification in home, further i use as less digital equipment as possible.
A warning when modding the UCD's, they do not like DC on the input when the Caps are removed or bypassed( UCD output goes pumping). When using tubes you need a delay at the input of the UCD so that the tube first needs to warmup before connecting the UCd to the tubestage.
Attachments
Last edited:
mods can mean breakdowns.
The story was that the FDP120N10 devices on the hypex UCD180HG....were built out of unobtanium, for the past few (3+) years. like being in a desert, none to be found anywhere.
Digikey now has a stock of them.
FDP120N10 Fairchild Semiconductor | FDP120N10-ND | DigiKey
Get them while you can.
The story was that the FDP120N10 devices on the hypex UCD180HG....were built out of unobtanium, for the past few (3+) years. like being in a desert, none to be found anywhere.
Digikey now has a stock of them.
FDP120N10 Fairchild Semiconductor | FDP120N10-ND | DigiKey
Get them while you can.
Last edited:
How much DC is there when you noticed the output pumping?
I like to DC couple my equipment, my DAC has max 30mV DC on the (differential balanced) outputs. There's also a few volts common mode on the outputs relative to ground though.
Hi Guys
Was wondering if anyone have use the aux supply of the Smps to power
the HxR directly. If so any improvement to the SQ ? Im thinking of trying
it, even wrote to Hypex but they can't advice on how to do bypass the supply
on the module
Thks
Was wondering if anyone have use the aux supply of the Smps to power
the HxR directly. If so any improvement to the SQ ? Im thinking of trying
it, even wrote to Hypex but they can't advice on how to do bypass the supply
on the module
Thks
- Status
- Not open for further replies.