Hypex or NewclassD? Which is the Best Class D Amp there is????

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
D

DocLorren

Member
Joined 2004
I switched from class A (DIY Death of Zen by ESP; overall a very nice low power transistor amp), to AB (DIY Gainclone LM3875; better bass, better dynamics) to class D (Sonic T-amp; great sparkle, voices/cymbals crystal clear) and finally the (DIY) AMP6 from 41Hz (same signature as Sonic T but even more of the goodies and lots of bass). This is my main amp now. And finally...yesterday I started with the prototype 10W class D monoblocks proposed one year ago by Brian W (BWRX). (Sorry Brian, I told you; terribly busy and so on. They will be great though!)
Shortly I will listen again to all amps (blinded! together with a buddy of mine to do some comparisons) but I have this feeling that class D is still surging ahead in sound quality while the more conservative topologies are more or less stationary. Doesn't keep me from trying out new amps all the time though (after the BWRX modules, I'll probably going to do the Pass F4).
All this btw. on Fostex 108EZ drivers in Buschhorns MkII without filters or damping.
 
soongsc

soongsc

Member
Joined 2005
BWRX said:


Some class d amps already out on the market have been given very favorable reviews compared to "state of the art" linear amps.
Click to expand...
I've listened to the TACT millenarian II and found it very clean with good soundstage image width and depth This is what I would consider as a class D reference. Having listened also to many linear amps such as Goldmund and MBL amps, I would say this amps are not as neutral as the TACTs, but have their own pleasing sonic signatures that make them easy to fall in love with. The NCD I have has some sonic features that are also quite nice, not as clean and precise as the TACT.
 
R

rogs

Member
Joined 2006
Ouroboros said:
I would say that Hypex still lead the field, but I do have concerns about the short-circuit protection of their amps.
Click to expand...


Any special reason ? I have not been able to blow up my UCD180's yet, and I have been trying quite hard (evaluation for possible commercial use). From my experiments with the Philips circuit, it soon became apparent that you need both overvolt and overcurrent protection, but I've found the UCDs seem to shut down OK with a short applied- even under heavy load conditions.
 
Ouroboros

Ouroboros

Member
Joined 2003
rogs said:



Any special reason ? I have not been able to blow up my UCD180's yet, and I have been trying quite hard (evaluation for possible commercial use). From my experiments with the Philips circuit, it soon became apparent that you need both overvolt and overcurrent protection, but I've found the UCDs seem to shut down OK with a short applied- even under heavy load conditions.
Click to expand...


I'm designing both a 60W and a 150w Class-D amplifier for the company I work for, which will be part of a 'voice-alarm' / PA application. This amp will feed a 100v line speaker system via a step-up transformer and so I need post-filter (but pre-transformer!) feedback to keep the required frequency response over a wide range of load impedances. I've had a bit of trouble in getting a short-circuit protection system that fully protects the output FETs when a short circuit is applied to the outputs when the amp is playing at high output volume. I initially tried using one of my UCD400 units as a reference test unit (feeding the 100v step-up transformer) and it killed the ouput FETs as soon as I short-circuited the output.
I don't know if the UCD amps implement proper cycle-by-cycle current limiting. Certainly the original Philips reference design shows a current limit circuit that is hopelessly inadequate.
 
Ouroboros

Ouroboros

Member
Joined 2003
Yes, I bread-boarded the sample I have of the DIL module onto a pcb. The amplifier works very well indeed into loudspeaker loads, but it had problems when driving the step-up output transformer to drive the 100V loudspeaker line.

I've just checked the Sequoia website, and the links to the data sheets do seem to be broken. I have the Sanyo data, but perhaps it might be available directly from the Sanyo website. I don't know what the nature of the business agreement is between B&O and Sanyo.
 
M

Midiup01

Member
Joined 2007
Class D - Sanyo

I found very interesting the product / hybrids from Sanyo.

I am looking to the STK428-640, used in some JVC equipments, but I cannot find the datasheets or the tech notes, does someone knows I to find it?

The Sanyo web site has only some brief infos and the distributor Sequoia shows just a preliminary presentation.

Thank you!
 
Ouroboros

Ouroboros

Member
Joined 2003
I've got all the data from when I was looking into these (November last year). It's in a 7 Meg Zip file. Do you want me to email it to you? Drop me an email with your direct email address and I'll send it to you.

I've not seen else much about these chips and modules from Sanyo. Perhaps there's some business or technical issues holding things up?
 
R

rogs

Member
Joined 2006
Ouroboros said:


I don't know if the UCD amps implement proper cycle-by-cycle current limiting. Certainly the original Philips reference design shows a current limit circuit that is hopelessly inadequate.
Click to expand...

I agree - if you build the Philips OCP as drawn, it wouldn't actually work at all !(D1 and D2 are shown reversed on the fig 9 schematic).
It also needs to be used in conjunction with OVP, otherwise the negative rail pumping justs destroys the output devices, with a short applied.

I think the Hypex amps (well the 180 anyway) uses just a current sensed OCP - I can see no reason for the 0.5 R Dale resistors otherwise. My versions also include Lm393 comparators, and a 4093 Schmitt NAND, which I suspect is used to trigger the amp 'off' function, when either over volts or over current are detected.

I have not been able to destroy my UCD 180s, even when applying a short across the secondary of a 100V line transformer load, under full signal conditions- although that is only with +/- 28V rails used.

I did manage to destroy the output devices in my own version (based on the Philips circuit, but with differential feedback, and both OCP and OVP fitted), but only if I used the recommended Philips devices. If I use IRF520 output devices, with their higher Rds on resistance, it seemed OK.

It seems strange that Bruno would have allowed such an 'iffy' protection system to be fitted to the UCD 400, when he has clearly taken such care with the rest of the design?

Although, as I say, I have had no problems with the 180 (so far!!)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom