• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Hypex UcD180HG HxR amplifier module analysis and review

OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
I wonder if you have seen similar, or would want to experiment with this kind of load. And it would be interesting to verify that Class D with PFFB have no problem with this (unfortunately I don't have a hypex or purifi amp to try).

Yes, I have a dummy load with similar impedance plot with negative phase above 5kHz and the A07 amplifier has similar FR issue with it just as you have shown. UcD and NC do not have this issue. As I have mentioned in post #1, the real reason is in different circuit principle of UcD and NC amplifiers compared to hysteresis switching principle (TPA3XXX)

It is a self-oscillating circuit with phase-shift control using the reconstruction filter L1,C1 and phase lead network R3,C2. The frequency of the self oscillations would be at the point where loop gain phase shift is 180°, see plots in Fig.3.

PFFB itself added to TPA3XXX does not help much to fix the issue. It is only a partial cure. A total cure is UcD/NC/Purifi topology with feedback taken only from the load and not in the mix from the load and in front of the LC filter, as is done in TPA3XXX PFFB amplifiers.
 

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
Yes, I have a dummy load with similar impedance plot with negative phase above 5kHz and the A07 amplifier has similar FR issue with it just as you have shown. UcD and NC do not have this issue. As I have mentioned in post #1, the real reason is in different circuit principle of UcD and NC amplifiers compared to hysteresis switching principle (TPA3XXX)



PFFB itself added to TPA3XXX does not help much to fix the issue. It is only a partial cure. A total cure is UcD/NC/Purifi topology with feedback taken only from the load and not in the mix from the load and in front of the LC filter, as is done in TPA3XXX PFFB amplifiers.
Thanks. I knew your dummy load was showing a problem but I wasn't sure how reactive your load was at high freqs [edit: I found the graph of your dummy load on Amir's dummy load thread; it is indeed very similar to my problematic speaker]. It is good to get a sense of what type of impedance causes the worst behavior, because you can build a decent system with an amp such as your A07 if you are aware of it and choose a speaker with consideration of its impedance characteristics.

Also interesting what you say about the TPA3XXX and PFFB. It would be interesting for someone with a Topping PA5 to do this kind of check with a reactive load to see how well the PFFB tames the peaking, or not.
 
Last edited:

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
Thanks. I knew your dummy load was showing a problem but I wasn't sure how reactive your load was at high freqs. It is good to get a sense of what type of impedance causes the worst behavior, because you can build a decent system with an amp such as your A07 if you are aware of it and choose a speaker with consideration of its impedance characteristics.

Also interesting what you say about the TPA3XXX and PFFB. It would be interesting for someone with a Topping PA5 to do this kind of check with a reactive load to see how well the PFFB tames the peaking, or not.
Sorry if I misunderstood something.

Many Class-D amplifier cards from example Ebay are based on the same principle and had a filter on the output, A coil. The weakness with these is that the frequency response changes with different OHM. load. same as a coil connected in series on a speaker does.

Remember that a speaker has different OHM. at different frequencies and do not have constant OHM.

HYPEX has solved this but patented this.
 

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
Sorry if I misunderstood something.

Many Class-D amplifier cards from example Ebay are based on the same principle and had a filter on the output, A coil. The weakness with these is that the frequency response changes with different OHM. load. same as a coil connected in series on a speaker does.

Remember that a speaker has different OHM. at different frequencies and do not have constant OHM.

HYPEX has solved this but patented this.
What I was trying to get at is that phase of the load seems to have more to do with creating problems than the impedance magnitude. My Sonos amp seems to behave nicely with a purely resistive load anywhere between 4 ohms and open circuit, but a load with negative phase angle above 10KHz creates the typical ugly FR peak.
 

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
What I was trying to get at is that phase of the load seems to have more to do with creating problems than the impedance magnitude. My Sonos amp seems to behave nicely with a purely resistive load anywhere between 4 ohms and open circuit, but a load with negative phase angle above 10KHz creates the typical ugly FR peak.
Ok. is Sonos amp
Class-D or Class-AB?
 

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
Sonos Amp is another Class D. And I really like it. It just doesn't work well with the one speaker I own with this type of impedance. All my other speakers work great with it.
 

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
Sonos Amp is another Class D. And I really like it. It just doesn't work well with the one speaker I own with this type of impedance. All my other speakers work great with it.
Ok. From personal experience after buying many Class-d amplifiers from ebay, there are a little too many who do not like low OHM. as 4 ohms and lower. Almost all are also made the same, the same basic construction.

Another thing I want to mention. From experience, cheap class-ab has better sound quality than the cheap class-d has. all cheap class-d amplifier cards i have purchased from ebay have poor sound quality. I have bought quite a few different ones.

from experience and what I have seen on youtube, L12-2 amp from ebay has good sound.
 

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
I’m pretty sure the specs in the datasheet are correct. Hypex is quite precise with this.

Something like 200VA should be fine (per channel)
Hello. Has now finally finished with amplifier construction of Hipex UCD180 ST.
I use a 45V 500WATT power supply which is 11.111A.
I use 2x 6A slow fuse. With 2x4ohm constant resistance loa,. fuse does not blow at max watts from both amplifiers. No volt drop either.

How many watts do I get out of the amplifier?
180w as hyupex writes? NO. Far away!
If I turn the volumes too high, the sound disappears until I turn down the volume! but volts from power supply still measure 2x45v.
I use ocilocope. I measure 100w before the sine curve out of the amplifier becomes too noisy.

The problem is that I measure 2 frequencies! a very high one and 50hz which I put in. Therefore, the sine curve becomes noisy with high frequency noise. My ocilocope can not filter out high frequency.

Can anyone confirm that the Hypex UCD180 has this protection feature that it turns off if you turn the volume up too high ??
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,223
Likes
17,800
Location
Netherlands
How exactly did you measure the 100W? For the 180W into 4R, you’ll only need about 27V.
 

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
Can anyone confirm that the Hypex UCD180 has this protection feature that it turns off if you turn the volume up too high ??
Mine behaves this way but I do not know if there is something wrong with mine or if it is made this way ??
It is not power supply! It does not drop under load!
 

Attachments

  • IMG_20220429_023255.jpg
    IMG_20220429_023255.jpg
    212.4 KB · Views: 163
  • IMG_20220429_023303.jpg
    IMG_20220429_023303.jpg
    238.2 KB · Views: 184
  • IMG_20220429_023314.jpg
    IMG_20220429_023314.jpg
    131.8 KB · Views: 161
  • IMG_20220526_193830.jpg
    IMG_20220526_193830.jpg
    194.5 KB · Views: 142
  • IMG_20220526_193927.jpg
    IMG_20220526_193927.jpg
    180.9 KB · Views: 164

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,223
Likes
17,800
Location
Netherlands
It says in the datasheet it has “Overcurrent and overvoltage protection”. Your 2x45V transformer is the DC voltage I hope?

Also see:
 
Last edited:

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
How exactly did you measure the 100W? For the 180W into 4R, you’ll only need about 27V.
I use the exact same power supply for my Naim nap250 clone board class-ab to measure a very nice and clean sine curve.

The switching frequency (f) for most Class D amplifiers is typically between 250kHz to 1.5MHz.

With the Hypex UCD180 ST and my cheap ocilocope I measure high frequency noise. possible this is switching frequency to the hypex amplifier. therefore, it is a bit difficult to see when the amplifier is cutting. I use this equipment. see pictures.

I use different sine frequencies and a constant 4 ohm resistor.

Remember many years ago I used a car bass to test hypex. did not get more than a little over 100watt! The problem with a car bass is that ohms are not constant throughout all frequencies
 

Attachments

  • IMG_20220529_164312.jpg
    IMG_20220529_164312.jpg
    199.7 KB · Views: 83

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
It says in the datasheet it has “Overcurrent and overvoltage protection”. Your 2x45V transformer is the DC voltage I hope?

Also see:
Yes of course! If you look at pictures of my amplifier, you will understand.
-- 45v-0v-45v + DC

Another thing I did was replace all the capacitors. The two amplifier cards did not have the same capacitors.

It was sanyo low ESR. I chose. A new one measured 0.35 ohms while the old one almost 2ohm.
 
Last edited:

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
It says in the datasheet it has “Overcurrent and overvoltage protection”. Your 2x45V transformer is the DC voltage I hope?

Also see:
Hello again. Has now managed to measure hypex ucd 180 st in a proper way.
what is written in data sheet of hypex ucd180 is not correct!
(180w / 4ohm, 105w / 8ohm)

Just before the amplifier turned off (over current protection)
I measure the following with 80hz sine tone.
23V 5.92A = 136.16 watts
The hypex ucd180 does not cut in the usual way as you are used to from many other amplifiers but you get more and more noise at the top of the sine curve. maybe soft cliping or similar.

Should you have a 100% pure signal without noise, you will not get more than 100watt out of the hypex ucd180

Is my power supply too small? No, I do not measure any voltage drop.

Noise filter coil at the output of the hypex ucd180 card gets relatively vrm! why? I'm guessing it's going to be 40 degrees. far from being destroyed by heat but still quite warm
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,223
Likes
17,800
Location
Netherlands
Possibly @pma has tested with a bigger power supply and can attest to the max power output into 4R?
 

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
Hello again.

Has now bought Hypex UCD180 HG and upgraded from ST.
UCD180 HG is marked with the year 2022.
I'm a little disappointed with one thing.
Hypex uses pretty rubbish low quality electrolyte capacitors from brands one should not use.

1. samwha RD. (10UF100v) is Standard series for general purpose and has a bad reputation for being unreliable.

2. BC PHILIPS VISHAY 036 CAPACITORS. (22UF 63v)
BC Components may have bought Phillips '' lytics lines. Anyway, those are 2 decades old, and modern low impedance caps are far better than low impedance caps of that time.

3. koshin capacitor (220UF 16v) is Hong Kong / China company.
They are on the list of capacitors you would rather not use.

Do any of you have experience with this?
UCD180 HG (HxR) is not a cheap amplifier and you expect better from Hypex.
The older UCD180 ST marked with the year 2005 also did not have better capacitors.
They should have used Panasonic FC or similar.
 

Matias

Master Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
5,029
Likes
10,796
Location
São Paulo, Brazil
Hello again.

Has now bought Hypex UCD180 HG and upgraded from ST.
UCD180 HG is marked with the year 2022.
I'm a little disappointed with one thing.
Hypex uses pretty rubbish low quality electrolyte capacitors from brands one should not use.

1. samwha RD. (10UF100v) is Standard series for general purpose and has a bad reputation for being unreliable.

2. BC PHILIPS VISHAY 036 CAPACITORS. (22UF 63v)
BC Components may have bought Phillips '' lytics lines. Anyway, those are 2 decades old, and modern low impedance caps are far better than low impedance caps of that time.

3. koshin capacitor (220UF 16v) is Hong Kong / China company.
They are on the list of capacitors you would rather not use.

Do any of you have experience with this?
UCD180 HG (HxR) is not a cheap amplifier and you expect better from Hypex.
The older UCD180 ST marked with the year 2005 also did not have better capacitors.
They should have used Panasonic FC or similar.
It has been discussed here ad nauseam.

 

opel

Member
Joined
Dec 30, 2021
Messages
28
Likes
0
It has been discussed here ad nauseam.

Ok. Thanks. This thread is closed. is that right? why?
 

tmtomh

Major Contributor
Forum Donor
Joined
Aug 14, 2018
Messages
2,636
Likes
7,487
Ok. Thanks. This thread is closed. is that right? why?

The person who was moderator at the time explains the reason in the final post of the thread.
 

arvidb

Member
Joined
Dec 1, 2022
Messages
96
Likes
114
Location
Sweden
I happened to order a pair of Elac DBR62:s now that they were on sale, and am also realising I could use a new amp and so I am looking at options.

Assuming 20 % VAT, a pair of UcD180HG HxR plus the SMPS400A180 incl cable sets, plus let's modestly assume 100 € for a case and connectors - and we're looking at over 400 € for a kit that will require drilling, soldering and assembly. Compare that to Audiophonics' MPA-S125NC that's 399 €, or the MPA-S250NC that's 499 €. And those nCore amps should give much better SINAD and even lower idle noise.

Is there still anything going for these UcD180HG modules? I can think of two possible things: you get to choose the form factor of the enclosure (the very deep cases of the MPA line can be problematic, perhaps), and if you need three channels but still less than 400 W of power in total these modules may make sense?

Is that a good conclusion?
 
Top Bottom