UCD180 questions

Jump to Latest
E

ewildgoose

Member
Joined 2004
Bruno Putzeys said:

(The new "esoteric" design, 1.2kW/2ohms, on the other hand, uses 2x20000uF all black gate caps and a 1kVA transformer. Yeah it sounds a bit better :).
Click to expand...

I'm very intrigued by the UCD modules, but the low-ish power handling is the biggest thing in the way from ordering one of the 180 modules.

Are you intending to offer this "esoteric" design for sale through Hypex?

Also, could JP comment on what kind of time frame before the 400 modules will be available for sale (hints on the board here suggest that they might be obtainable right now if you ask nicely?)

Thanks
 
R

rosbacke

Member
Joined 2004
rosbacke.org
Pruno Putzeys
Yes/no. In general MOSFET switching times increase with increasing voltage (more gate charge to throw about). It means that the relative portion taken up by dead time gets longer at lower voltages. If an amp was designed to operate at 50V, reducing supplies to 25V, THD is likely to go up. The UcD180 module is best used as designed.
The UCD400 circuit has drivers that prevent this mechanism (the reason for this design change was different though). It will produce the pretty much the same THD at any power supply voltage. Of course, who would buy a UCD400 module to run it at 25V...
Now, if you only need 10W at any time, you're better off building a power stage with lower voltage FETs (same Ron), because it'll easily provide faster switching times than a similar power stage made with higher voltage FETs.
Click to expand...

Thank you Buno. It is a good answer.
My understanding is that when using Ucd180 it is a good thing to let it operate att +-50V all the time. If I want to use less voltage, I'm not operating the module under optimal conditions.
I have two 22V/160VA twin coil transformers which I can choose to operate in a serial or parallell mode. I serial, Ihave to use a linear regulator to shave of a couple of volts since I get +-60V without a load. The added benefit of lower power ripple makes it attractive even though there are some added components so I'll stick to that plan. If anybody is interested I can post the schematics when it is finished.

--- Mikael R
 
peranders

peranders

Member
Joined 2002
sjostromaudio.com
Paid Member
ewildgoose said:


I'm very intrigued by the UCD modules, but the low-ish power handling is the biggest thing in the way from ordering one of the 180 modules.

Are you intending to offer this "esoteric" design for sale through Hypex?

Also, could JP comment on what kind of time frame before the 400 modules will be available for sale (hints on the board here suggest that they might be obtainable right now if you ask nicely?)
Click to expand...



Jan-Peter, Bruno, maybe this could be an idea, to let some people here have promotion copies (to attractive prices) of your new 400 watt module let those people be like beta testers? I'm sure you would get very useful input of many engineers and non-engineers. :idea:
 
IVX

IVX

Member
Joined 2003
e1dashz.wixsite.com/index
Bruno,
The main reason is 2x20000uf or BG? I remember you told that usual el.cap+ceramic is better then BG or so? :scratch:

Mr Teal,
<car audio subs tend to be designed with large motors, big VCs, and stiff suspensions, with makes them great for providing loads of output in very small boxes.>
Therefore carrier based classical PWM with feedback before filter (output impedance .04ohm and THD 2%) have uncontrolled bass. :dead:
 
B

Bruno Putzeys

Member
Joined 2004
hypex.nl
IVX said:
The main reason is 2x20000uf or BG?
Click to expand...
I tend to think it's largely because of BG...
IVX said:
I remember you told that usual el.cap+ceramic is better then BG or so? :scratch:
Click to expand...
I can't possibly have said such a thing. Parallelling elcaps and ceramics is not to be done lightly. You always need one (or more!) smaller elcaps to damp the parasitic LC circuit that results from the big elcap and the small ceramic.

So far I've found that the only capacitor capable of working magic close to BG's is a mix of Elna Cerafines and BC Components "035". Unfortunately the latter are no longer being made.
 
B

Bruno Putzeys

Member
Joined 2004
hypex.nl
matjans said:
have you tried panasonic fc? they work very well with chipamps/gainclones.
/matti
Click to expand...
Not this particular kind, but the panasonic caps I've tried (various kinds) always sound a bit thin in the mid-range. If the amp is a bit "squawky" and lacks HF presence, panasonic caps help, but usually they just come out as thin and putting too much sheen on the top end.
 
S

Saam

Member
Joined 2004
low power UCD modules

Bruno Putzeys said:


Now, if you only need 10W at any time, you're better off building a power stage with lower voltage FETs (same Ron), because it'll easily provide faster switching times than a similar power stage made with higher voltage FETs.
Click to expand...

Hello all,
I asked some time ago to Jan-Peter to get low power modules. It seemed only possible to lower the voltage. Now you write that to get lower power, one should use lower voltage FETs. Is it possible to get it from Hypex?
If not, is it 'only' a matter of changing the FETs on the UCD board? Can it be done easily? Do you have any suggestions regarding the FETs that can be used?

The UCD module is very interesting for a use in active speakers, but 100W is very often a waiste of power in this configuration so I think many people might be interested in lower power modules.

Thanks

Olivier
 
G

ghemink

Member
Joined 2004
Bruno Putzeys said:

Yay!:)

The modules have about 65dB PSRR (frequency independent). I tend not to use overrated supplies because I found the sonic benefit to be rather small. The quality of the power supply caps does more than the quantity.
My own reference amp has a 160VA transformer and 2x4700uF - for a stereo amp (2x90W).
(The new "esoteric" design, 1.2kW/2ohms, on the other hand, uses 2x20000uF all black gate caps and a 1kVA transformer. Yeah it sounds a bit better :).
Click to expand...


Hi Bruno,

I have a whole bunch BG caps that I want to use in the UCD180. For all caps I have an equivalent BG (all of them powerrail decoupling and coupling caps between opamp and UCD). However, I don`t have an equivalent for the 100uF cap that is (I suspect) used for the bootstrapping of the gate of the power NMOS that is connected to the positive rail. I guess that CAP is less critical for the sound quality, can you confirm that? Could that CAP be replaced with for example a 22uF BG? I don`t really want to experiment on my own with that cap since it is playing a vital role in the power stage, so advice is appreciated.

By the way, those 470uF caps that are placed between power rails and ground before the two small resistors (current monitoring resistors I think), can they be replaced with say 100uF BG caps? I think so, especially when the wiring from powersuplly to the modules is short. On the other hand, I think those 470uF caps are less critical, probably the 2x22uF caps that are used close to the power NMOS devices are more important for the sound quality.

Best regards

Gertjan
 
B

Bruno Putzeys

Member
Joined 2004
hypex.nl
Re: low power UCD modules

Saam said:

Now you write that to get lower power, one should use lower voltage FETs. Is it possible to get it from Hypex?
If not, is it 'only' a matter of changing the FETs on the UCD board? Can it be done easily? Do you have any suggestions regarding the FETs that can be used?
Click to expand...
I wouldn't just start swapping out mosfets on an existing amp. Unless you can get at the right types you might end up with worse performance.
If JP sees business potential in a low-power module he may decide to make a product, but that remains to be seen.

In the meantime, just use the 180W module at normal voltage. There is no power "wasted" - remember it's class D. Making a lower power module would only shave off a watt or so (idling losses). If you're using the module at lower power, you can simply use a smaller transformer, but with the same voltage rating.
 
J

Jan-Peter

R.I.P.
Joined 2003
hypex.nl
Hi,

We are busy with a new manual for the UcD Modules, in about two weeks this be available on our website. (sorry, busy, busy.....)

Per-Anders,

Thanks for your idee, but we have already a lot of volunteers ;)

Gertjan,

You don't will get any advance with replacing the 100uF capacitor to BG. This has no influence on the sonic.

In a few weeks we have available for the UcD180 a FULL toroidel powertransformers where also the crossection is round.

Regards,

Jan-Peter

www.hypex.nl
 
B

Bruno Putzeys

Member
Joined 2004
hypex.nl
ghemink said:

I have a whole bunch BG caps that I want to use in the UCD180. For all caps I have an equivalent BG (all of them powerrail decoupling and coupling caps between opamp and UCD). However, I don`t have an equivalent for the 100uF cap that is (I suspect) used for the bootstrapping of the gate of the power NMOS that is connected to the positive rail. I guess that CAP is less critical for the sound quality, can you confirm that? Could that CAP be replaced with for example a 22uF BG? I don`t really want to experiment on my own with that cap since it is playing a vital role in the power stage, so advice is appreciated.

By the way, those 470uF caps that are placed between power rails and ground before the two small resistors (current monitoring resistors I think), can they be replaced with say 100uF BG caps? I think so, especially when the wiring from powersuplly to the modules is short. On the other hand, I think those 470uF caps are less critical, probably the 2x22uF caps that are used close to the power NMOS devices are more important for the sound quality.
Click to expand...


BG caps are big for their value, aren't they?:D

The boot strap cap does have an effect on sound, but it should not be less than 100uF. In self-oscillating amps there's no recharge during clipping, so the cap should hold enough energy to keep the gate driver happy during that time.

The 22uF's are a bit less important to sound than the 470uF caps. I would not recommend changing the 470uF caps to 100uF, for reasons of ripple current. The whole point of these caps is not to provide storage, but to keep the switching ripple current local to the amplifier.

In any case, I found the pre gain stage is the most important factor in sound. All the rest comes next.

Cheers,

Bruno
 
You must log in or register to reply here.
Top Bottom