• Welcome to ASR. 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!

NEW BRZHifi PAD-30 : Dual TPA3255 +PFFB @ $99

You might be right (something to check); in that case, I’d be forced to save about a dollar and the time it takes to make two solder joints—what an ordeal... :eek:
Got an opinion from a colleague on them. He does PCB design and sources parts. Said those Rubycon ZLH are not a bad choice. Can go above 470 a bit there, but not much as it can create other problems. 1080 is def. too much. When upgrading and changing better look for prolonged life specs. 4k hours -> 8k hours spec choices. I hope to get some recommendations soon. Depends on what is available (EU here).
 
Hello Sam_138, and thanks for your contribution.

Could you also provide the height of these capacitors (normally 20mm)—or at least the maximum height available to avoid touching the case (measured from the PCB surface to the underside of the top cover)?

TonyJZX confirms—as I mentioned earlier—that these are good capacitor models; however, referring to the Texas Instruments datasheets, the value chosen by the manufacturer is too low: 4 capacitors in parallel per TPA3255 (there are 8 on the PCB) give a total of 400µF.

The datasheets show that two 470µF capacitors are used for each TPA3255 (totaling 940µF) when the TPA3255 chips are operated in PBTL mode ->

View attachment 544727

Therefore, to operate according to Texas Instruments' specifications, it is necessary to replace the capacitors on the PCB.

There are other Rubycon capacitor models with better specifications that allow this result to be achieved; I have used the ZLH Series models as an example ->

View attachment 544729

You will notice that I have selected 63V models, which are more than sufficient for the PVDD voltage being used (the amplifier manufacturer recommends a maximum power supply voltage of 48V).

A 10mm diameter was chosen because it fits perfectly into the original PCB mounting locations.

The individual capacitance values will be either 270µF (4 x 270µF = 1080µF) with a height of 20mm—matching the originals—or 330µF (4 x 330µF = 1320µF), provided the 23mm height fits within the enclosure.

Replacing the capacitors in accordance with the datasheet recommendations will improve the amplifier's response to strong signal transients and reduce the load on the power supply.

I will return later to the subject of selecting a power supply for this amplifier, as the manufacturer's power chart can be misleading: it lists the power output for a single TPA3255 chip, whereas the amplifier actually contains two. To run it at full capacity, you therefore need a power supply capable of delivering double the current (in amperes) specified!

However, this requires careful consideration of how the amplifier will be used, taking into account both its heat dissipation capabilities and the level of distortion acceptable for hi-fi listening...
Note: The Rubycon ZLJ series also exists and would be suitable, but it is quite often out of stock.

Good morning from Spain,

Yes, the length is 20mm and the available space is about 24mm, it would be a very tight fit.
 
there's been some time since this thread was started

i was cruising aliexp and i noticed there's a few sellers of this now... if you have a psu then you can buy the amp for well under $75 usd

i would assume you guys have a few loose psus lying around

could well be as low as it gets prior to 11/11

aiyima and other famous brand 48v 5a psu are like $25 and i assume they all use the same plug
 
@ Ruediger: 270µF RUBYCON ZLH are 10000 hours, juste ZLJ are better ((as I mentioned earlier, you need to take the time to observe carefully).
Like your friend, I was involved in PCB design for a brand that is now well-known; we simply have a different perspective, that's all ;)
@Sam_138: Once again, thank you for the details you provided for me and the other readers.
As you may have read above, I chose 270µF capacitors with the same height as the originals installed on the PCB; I suspected space would be limited, and components need to 'breathe' :)
@ TonyJZX: Regarding the power supply to use with this amplifier, here is my take:
- The 48V 5A power supply won't be 'powerful' enough to properly drive the two TPA3255 chips during high power demands (signals modulated in the lower part of the audible spectrum).
- The 48V 10A power supply might work but seems ill-suited to me, as the amplifier will run too hot at idle, and the maximum power output exceeds my distortion limits (excessive THD+N); in short, it doesn't seem like a good fit either.
So, based on my criteria (see above), I chose a power supply that suits my needs ->
AUDIOPHONICS - PSU 42VDC - 8.5A.jpg


You will read above, with some details, that I specified this is a model distributed by the Cisco brand.
It seems Cisco has been around for a very long time and, given the nature of their business, is required to provide high-performance, reliable equipment, right?

I won't go into the details of what motivated my choice here; instead, I invite you to check out the thread on the AMPAPA D1 amplifier, where I discuss factors such as power, THD, and heat dissipation...
 
- The 48V 10A power supply might work but seems ill-suited to me, as the amplifier will run too hot at idle, and the maximum power output exceeds my distortion limits (excessive THD+N); in short, it doesn't seem like a good fit either.
Can you elaborate on the too hot part? is that specific to this amp?

The 48v10a GaN brick seems standard for these 2xTPA3255 amps as it matches their max. output roughly.

That no one actually wants to draw 2 x 240w with a load of distortion is another thing.
 
Idle power draw depends on supply voltage. I don't see any dwnside of more current, except of price.
 
  • Like
Reactions: MCH
I don't get the logic in selecting a lower voltage supply. The amp distorts because it reaches the voltage limit at 280 W (about 33,5 V * 1.41 = 47,2 V @ 8.4 A into 4 Ohms). If you select a lower voltage supply, you will simply reach that limit earlier.
 
Your point is valid, popej, as you highlight that it depended on the voltage—which is indeed the case for the idle state with this type of amplifier.

Moving beyond purely theoretical considerations, I encourage anyone who doubts my claims to experiment—as I have done many times—with various amplifier models of this type using different supply voltages (24V, 32V, 36V, 38V—as with the Topping M300—42V, which I chose, and 48V).

You can then decide for yourself what you consider acceptable.

Randomear: your reasoning is sound, but it doesn't reflect the reality of my situation because the amplifier uses two TPA3255 chips, I am using 8-ohm speakers, and I am factoring in the distortion levels associated with specific power outputs—data visible on many measurement charts and curves.

You will find that, to stay within acceptable distortion limits for a "Hi-Fi" setup, that limit is actually reached at lower power levels (around 70W to 100W RMS per channel seems to be the maximum for acceptable distortion—well before clipping or even hitting 1% THD).

Furthermore, even 100W RMS—driving speakers with, say, 90dB/W/m sensitivity—would yield a sound pressure level of 110 dB SPL at 1 m, so approximately 100 dB SPL at 3 m.

That is substantial, especially considering it is far from certain that the speakers themselves could even handle such a level.

You need only look at most of the speaker reviews measured by Amirm, which show that, generally speaking, they can handle a maximum level of around 96 dB SPL...

That said, I’ll turn the question back to you: why choose a more expensive power supply—one that will raise your amplifier's idle temperature—when you will never actually use the maximum power it provides?

Personally, I see no advantage to that; I prefer to trust Cisco and its reputation (in data centers, for instance) over Aiyima or Fosi, even though the latter do make decent power supplies.

As I always say: everyone should make their own choice, provided they are happy with the result.
 
- The 48V 5A power supply won't be 'powerful' enough to properly drive the two TPA3255 chips during high power demands (signals modulated in the lower part of the audible spectrum).
- The 48V 10A power supply might work but seems ill-suited to me, as the amplifier will run too hot at idle, and the maximum power output exceeds my distortion limits (excessive THD+N); in short, it doesn't seem like a good fit either.
The question (that you didn't answer) was:

Why should the amp run hotter in idle with a 48V 10A PSU than with a 48V 5A PSU?
 
Received my unit.

It comes with a cheap 48v5a psu and a cable - 3-prong into the psu but 2-prong into the socket. Not sure what to make of that (error?). Psu has 3-prong input.
 
hharkpabst: You are misinterpreting what I said; I wrote that the 48V 5A unit wasn't powerful enough for the two TPA3255 chips, and also that the 48V 10A unit caused more heating while idling.

In other words, the point isn't that the 48V 10A unit runs hotter than the 48V 5A unit, but rather that neither is "suitable" (in my opinion): one lacks sufficient power, while the other has enough power but generates excessive heat regardless. The key takeaway is that the 48V supply itself is the "cause" of increased heating in the TPA3255 chips during idle—even though it is powerful enough to drive them "adequately" if the chassis's heat dissipation allowed for "intensive" use (near maximum power), assuming the speakers could handle it.

We simply need to explain the concept of power and its implications to readers: many are looking for high power for a specific purpose (which needs to be defined), yet with this type of device, high power is only achieved for very short bursts (often milliseconds) rather than continuous operation. This isn't specified in the manufacturer's power chart for this amplifier, and furthermore, the figures refer to a single channel, whereas the amplifier has two.

Therefore, one cannot rely on these figures to correctly determine the final power supply to use with this amplifier.

I am trying to "demystify" the actual power output achievable with such an amplifier under optimal conditions, but perhaps there is a language barrier, or my points are poorly phrased or misunderstood; if an English-speaking member could explain this instead, it might be clearer.

@ Rudieger: If the PSU is 'Class II', then there is no issue; however, if that is not the case (meaning a live conductor—phase—could come into contact with the casing, which in turn could touch the user), then the PSU's power cable should include a third conductor for a ground connection at the plug end.

Most of the time, these power supplies feature an external plastic casing and are (typically) Class II devices; the connector found on the PSU itself is simply a manufacturer's choice based on IEC standards—most commonly the IEC C13 (3-pin) or IEC C7 (2-pin) type.

In short, for this type of Class II power supply with a plastic housing, an IEC C7 connector would suffice, even with internal conductors having a cross-section of 0.75 mm² or 1 mm².
NB: my 'guidelines' apply to countries like mine (France) that use a 220/240V supply voltage for the switched-mode power supply (SMPS) unit since delivering the same power at 'only' 120V results in double the current, the conductor cross-section would need to be larger (refer to the applicable specifications and regulations for the country in question).
In France (and thus in Europe), the applicable standards are the 'NF' and 'CE' marks; these are very strict and subject to evolution, ensuring—when properly followed and applied—safe and reliable operation for the device and, above all, for its users.
I am not familiar with the standards in other countries, so I encourage you to consult them, ensure compliance, and verify that the device in question meets those requirements.
 
Last edited:
It's plastic of course. Like all the others, GaN or not.

I actually remember another 48v5a from 3e also having this cable. But on both ends. Whereas the 48v10a bricks have full 3-prong cables.
 
This thread is specifically for collecting reco's and sources for PSUs

 
Randomear: your reasoning is sound, but it doesn't reflect the reality of my situation because the amplifier uses two TPA3255 chips, I am using 8-ohm speakers, and I am factoring in the distortion levels associated with specific power outputs—data visible on many measurement charts and curves.

You will find that, to stay within acceptable distortion limits for a "Hi-Fi" setup, that limit is actually reached at lower power levels (around 70W to 100W RMS per channel seems to be the maximum for acceptable distortion—well before clipping or even hitting 1% THD).
I think you expect to stay below that "kink" in the distortion graphs where they shoot up around 115 W into 8 Ohms for TPA3255's, right? I tried to hint at the fact that by using 42 V instead of 48, you do not stay below that point in the power graph. It occurs at pretty precisely 90% of the driving voltage, from my understanding simply due to losses in the amplifier. That means for you, that "kink" in the graph starts at 90 W instead of 115. You do not avoid it, you simply lower it proportional with the supply voltage.

Furthermore, even 100W RMS—driving speakers with, say, 90dB/W/m sensitivity—would yield a sound pressure level of 110 dB SPL at 1 m, so approximately 100 dB SPL at 3 m.

That is substantial, especially considering it is far from certain that the speakers themselves could even handle such a level.

You need only look at most of the speaker reviews measured by Amirm, which show that, generally speaking, they can handle a maximum level of around 96 dB SPL...
I agree. In general, the difference between 90 W and 115 W will be irrelevant and with decently efficient speakers, both will push to uncomfortable levels.

That said, I’ll turn the question back to you: why choose a more expensive power supply—one that will raise your amplifier's idle temperature—when you will never actually use the maximum power it provides?

Personally, I see no advantage to that; I prefer to trust Cisco and its reputation (in data centers, for instance) over Aiyima or Fosi, even though the latter do make decent power supplies.

As I always say: everyone should make their own choice, provided they are happy with the result.
Simple: I want to have comfortable headroom for instantaneous, short peaks in the SPL and thereby power. I don't know if that PSU is from or for Cisco or which will last longer. It is not critical either way: The amp will work fine with 42 V and it will be a tiny bit cooler, yes.


Different topic: Please use the "Reply" function or mention members with "@". Otherwise, we will not get notified of your responses.
 
@RandomEar

Thank you for your input, as it aligns with what I was trying to convey ;)

In my case, I will never reach the 'inflection point'—which lies precisely at 90% of the supply voltage—because the sound pressure level would be unbearable; however, by using a 42V power supply, the unit does run slightly cooler (whereas my tests with the stereo AIYIMA A70 and a 48VDC supply resulted in an external case temperature that I considered excessive).

I thus achieve the goals I set for myself:

- power that is more than sufficient for my needs.
- operation with minimal distortion during listening sessions.
- a device that runs at a much more suitable temperature, likely ensuring better long-term reliability.
 
Last edited:
People hi :D

Amplifier received, so time to unbox:

1 - The BOX

1 - The BOX.jpg


2 -Box details

2 -Box details.jpg


3 - Inside of the box

3 - Inside of the box.jpg


4 - Front view

4 - Front view.jpg


5 - Rear view

5 - Rear view.jpg


6 - Side view

6 - Side view.jpg


7 - View from below (The orange arrows indicate the 4 screws for releasing the top cover)

7 - View from below.jpg


8 - Use a 2mm Allen key for removing the top cover (the 4 screws underneath)

8 - 2mm Allen key for removing the top cover (the 4 screws underneath).jpg


9 - My new selected OPAmps

9 - My new selected OPAmps.jpg


In the next post, I will briefly explain—using an image—the reasoning behind my choice of op-amps, although I will soon be replacing the OPA1612, bipolar type (on the right) with another OPA1642, JFET type, model on which TEXAS INTRUMENTS has particularly worked on their electromagnetic interference (EMI) rejection ratio, or EMIRR, (describes the EMI immunity of operationalamplifiers) which strikes me as a fairly important point regarding this type of amplifier and the everyday devices that surround us (smartphones, Bluetooth, switching power supplies, etc.) ;)

It doesn't matter whether you follow my choice or go with one that seems right to you, but—just as a side note—the op-amp located on the far right of the unit I received turned out to be "microphonic" (lol) :facepalm:

However, as you'll see in my next post, it's important to note that the positive and negative power rails for the op-amps operate at 5 volts (DC, of course).
Consequently, it is strongly advised to steer clear of so-called "Class A" op-amps (which often or always run on 12 volts) sold at exorbitant prices; this is actually a good thing, as they are unnecessary, often cause more distortion, and run extremely hot :confused:

At these voltage levels, they either won't work or—worse—will damage or destroy the circuitry by trying to compensate for the lack of "power" by drawing excessive current...
... you have been warned.
 
Follow-up to the previous post, with some useful technical information.

11 - PFFB circuit

11 - PFFB.jpg


12 - OC_ADJ & FREQ_ADJ resistors

12 - OC_ADJ & FREQ_ADJ.jpg


Visual explanation of the previous two photos ->

12.1 - BRZHIFI PAD-30 (PFFB, PWM and overcurrent).jpg


13 - One channel OPAmp

13 - One channel OPAmp.jpg


On this photo I wrote an annotation specifying that the OPAmp is used as a 'buffer', which is partly true because it does not 'produce' any gain, however it is partly false because the capacitor (framed in green) located on the feedback circuit of the operational amplifier will directly condition the shape of the 'downward curve' in the high frequencies in connection with the components of the PFFB circuits.

When the latter is poorly 'calibrated' then this is partly an explanation for the 'rise' in high frequencies when an impedance (for example 8 Ohms, rather than 4) as we see on amplifiers which do not have PFFB circuits or which are poorly 'implemented'.

(Lester from SYLPH Audio had explained and demonstrated this to me through simulations as well as with a simulator using BODE diagrams when designing his modified D200 MK II PCB :cool:).

14 - Line gain OPAmp (the one located next to the volume potentiometer)

14 - Line gain OPAmp.jpg


I haven't had time to reverse-engineer the PCB yet, so I can't give you any more information at the moment...

15 - OPAmps supply voltages (Important point raised in the previous post)

15 - OPAmps supply voltages.jpg


I also took temperature readings from the top of the case—using a thermometer positioned 20 cm away—with the two different PSUs I had available.

The ambient temperature was 27°C with 49% humidity for the following two readings:

- Idle temperature with 36V DC 6A PSU (recorded after 4 hours)

17 - Idle temperature with 36V DC 6A PSU (recorded after 4 hours).jpg


- Idle temperature with 42V DC 7.5A PSU (recorded after 4 hours)

16 - Idle temperature with 42V DC 7.5A PSU (recorded after 4 hours).jpg


With this same power supply and my AIYIMA A70 stereo amplifier, the temperature under exactly the same conditions was 39.8°C.

It can already be objectively stated that the BRZHIFI PAD-30 amplifier (with two TPA3255 chips) runs cooler than the stereo AIYIMA A70 (with a single TPA3255 chip).

Subjectively, I have nothing to say about this amplifier that would distinguish it in any "glaring" way from other amplifiers of this type.

Nevertheless, the sense of available power—as well as the impact (if one can call it that) in the bass register—is truly noticeable (despite using a "simple" 42V DC 7.5A PSU).

For instance, when listening at close range (at my desk) with my speakers and the volume set to the "10 o'clock" position, the perceived sound pressure is almost on the verge of being unbearable; so, in my case, my choice of PSU is more than "sufficient."

Note: I should mention that in a previous post, I spoke about a 42V DC 8.5A power supply available from Audiophonics, but the unit I actually received was different. I informed them of this discrepancy and mentioned the publicity I had given them here, suggesting they respond to the thread—though they haven't done so (yet?). Their reply was that they hadn't been notified of the change made by their supplier; they offered me the choice of either returning the unit or keeping it in exchange for a €10 voucher.

Here is my small contribution to this topic, pending objective measurements of this amplifier.

See you soon, and happy reading on this forum ;)

PS: @BRZHIFI ->
Could you make a DAC using the same chassis and dimensions—keeping it just as simple (no completely useless displays or LEDs, no MQA or Bluetooth)—featuring only RCA outputs, dual ES9039Q2M chips, 6 swappable op-amps for the audio output stage, a chip to handle a 10-band parametric EQ function, a robust and durable USB-B port, plus a simple "POWER" LED and a 3-position switch (ON/OFF/TRIGGER), all at an affordable price point so it pairs perfectly with this amplifier (I suggest the name DAD-30)?

With this amplifier (PAD-30), you really focused on the essentials, backed by what is clearly a masterful command of the technical aspects; a DAC sharing that same spirit would certainly be welcome—both for me and, I’m sure, for many of your other customers.

It would be a refreshing change from all those DACs loaded with complex features that are both useless and uninteresting... in addition to being exorbitantly expensive (justifiably so) :rolleyes:
 
Last edited:
Things I noticed:

Trigger port bad fit. Different cables, needs fiddling with every single one.

When triggered off the red power light blinks, very annoying.

The on/off volume wheel feels like it has a lot of play.

What I like is sound, power and temperature. Challenged it for some time, it performs very well.

It will now make it's way into my garage setup.
 
Honestly I think there are more important devices in the queue. Like the X3900H. ;)

With this little amp it's all about the spectacular price point for the package. And they even throw in a (cheap) 48v5a psu. But I don't expect it to be up there with the Vibelinks, A7s, etc.
 
Back
Top Bottom