Making of Lightforce 1 Class D Powerbriefcase

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

ghemink

Member
Joined 2004
KLe said:



Hi ghemink

Could you please calm down.:( Lars, in his post, was quite clear that the smps was intended for his ZP2.3SE modules. I believe that the smps should also work for the ZP700XE modules. Lars, is that correct?

Having said that, you can always try it with your Ucd modules, although Lars believes that it will not work as he intended. Please, let us know what the results are.:)

krgds
Click to expand...


Was I offensive????

I do not see any reason why an SMPS would not work with an Hypex amp so I would be very interested to know why not as Lars claims that it would be like a complete disaster. I have not heard any statement that explains why it would be a problem with UcD and why not with ZAP. So please explain why so that
we can all learn something instead of being bashed again for being non-ZAP followers.

Just to let you know, I have owned 4 ZAPpulse 2.2SE modules in the past and these did not work for me unfortunately, they sounded better than my Accuphase class AB amp, so they are not bad. They were just too noisy for my application(active speakers) the noise from the tweeter directly connected to the ZAP was too loud and clearly audible at the listening position. Of course I had other issues with them as well as you maybe able to find out when reading some of the threads out there. But I'm not here to bash the ZAPs as in the newer ZAPs the issues I had maybe solved.

Actually, I'm working with an SMPS for my UcD modules but the purpose is not to make something fetherlight, just wanted to have a regulated supply. I have added lots of caps at the output of that SMPS and have changed the feedback loop of the SMPS accordingly (to stabilize the feedback loop). As a result, a very stable supply voltage is obtained which gives good results with my UcD400 modules (now for woofers only, but later also for the mid and high amps). So I'm very interested in what the issues are when an SMPS is used with the UcD modules. I'm not interested in wild claims made by Lars that look just groundless to me. So please explain, maybe there is a good reason why the UcD is not suitable for an SMPS but I can just not see it.

Best regards

Gertjan
 
G

ghemink

Member
Joined 2004
Lars Clausen said:
ghemik: I am sorry if my above message can be misunderstood as bashing of your well established friends at hypex. It is not intended. But i really needed to point out the potentially misfortunate outcome, of combining my smps with their amplifiers, especially because the expected problems would normally be associated with power supply problems in general. In fact i can't even be sure their amplifier modules work stable and reliable with the smps that i am building.

So the whole situation could easily be misunderstood, and that is what i want to avoid. I would have made the exact same recommendations for any other amplifier, where i would foresee problems of this kind. But in this case, Yves Smplders asked about using hypex.

You know JP has taken every opportunity to bash my stuff in the past, so let's just say that this way i am trying to minimize the bashing in the future, by keeping our stuff separate. ;) But maybe you hypex fans should ask JP to design a smps, that is fit for his amplifiers, if it is of interest for you?

All the best from

Lars
Click to expand...


Excuse me? I'm not connected with hypex in anyway, not more or less with you than with them. I'm both a customer of LCaudio and Hypex. I have bought 4 ZAPpulses in the past and I have bought more UcDs as they worked out better for my application (active speakers). I'm just interested in good sound and something that works for my application. I don't mind what people like, ZAP or UcD or whatever else. Just interested in learning things. So I would like to understand why the UcD is not suitable for your SMPS. You claim that there can be a "potentially misfortunate outcome". That is an interesting claim so I would like to know why so that I can learn something useful.

Best regards

Gertjan
 
T

tlf9999

Disabled Account
Joined 2005
I agree with ghemink. It doesn't sound there is anything inherent tot he design of the amps or the smps that the smps wouldn't work with UcD, or any other amps for that matter.

Unless of course Lar can point us in the right direction, which would be a great learning experience for us all.

Simplying making a statement without substantiation isn't much help.
 
P

Pierre

Banned
Joined 2004
Well, all things clarified, let's return to our "work". Let's try to keep this forum free of business discussions.

Lars, if you have decided to share/develop the design of a SMPS with no commercial intentions in mind, then count on me (and I am sure that on many others) to help in all that we can. I am sure you can do it yourself alone and very well of course, but if we can be of any utility...

Best regards,
Pierre.
 
T

tlf9999

Disabled Account
Joined 2005
Lars Clausen said:
For obvious reasons i don't know anything about the power supply requirements of the UcD's, i'm not designing my smps for use with UcD's but for ZAP's. For starters i think each amplifier need a different set of low voltage supplies. ??

It's that simple. :cool:
Click to expand...


Sounds like you wanted to say that your SMPS may not work with UcD (but there still is a chance of it working, you just don't know - fair enough).

Your earlier writing made, with reference to "lacking dynamic perfomance, and a messy soundstage" made it sound like it wouldn't work with UcD (as in sure disasterous outcome, no chance in hell, etc.).

semantics, i guess.
 
J

Jan-Peter

R.I.P.
Joined 2003
hypex.nl
Lars and myself have mailed about this topic, the UcD will defiantly work very good with a SMPS!

Last week I got an email from an OEM customer who has tested an UcD and an-other-very-famous-Danish-Class-D product (no it was NOT Zappulse!), the UcD performed slightly better and was more stable on this particuler SMPS.... ;)

Please go on with the discussion about the SMPS! We will NOT sell a SMPS to the DIY, so I would encourage everybody to help Lars to design a proper SMPS for an half bridge Class-D amplifier.

Gertjan has also done some interesting work, please see also his thread!

Last but not least, one of our other OEM customer use an UcD module together with an own designed SMPS, we were all impressed with the sonic performance of this combination. So I guess that all good sounding amplifiers (UcD, Zapppulse, Analog amplifiers), can have benefit with a good SMPS design!

Jan-Peter
 
Y

Yves Smolders

Member
Joined 2005
Jan-peter,

I'm sure you don't want to sell the SMPS to DIY because of the dangers of SMPS design?

I'm interested in SMPS in a commercial way - my company would like to equip some of our HTPC solutions with 5 channel integrated amplification - a regular PS would be hard to integrate into a PC case, SMPS won't be a problem.

I hope Lars will still work on his PS, if it is not complex it might be my first step into true DIY (unlike assembling some modules and simple parts like I did so far)

Here is a very elaborate document on SMPS design around a national semiconductor design:

LM5030 push-pull
 
KLe

KLe

Banned
Joined 2005
Lars Clausen said:
To keep weight and size down, a small but powerful switchmode power supply is used. (Schematic will be posted later).

The use of a switchmode power supply means the power can be easily upgraded. Since power output is mainly limited by the voltage of the power caps. (63V). You can use 80V caps, and put a few extra windings on the transformer to get 350 Watts per channel. (in 8 Ohms). No problem. I just thought 200 W per channel is more than adequate for my 3rd story downtown appartment with neighbors. ;)

I made the switchmode transformer myself, and it is wound with 15 windings on the primary side, and 6 on the secondary side. In between glass wick tape, to get a good safety distance.
But quite easy to build!

It's running 250 kHz, on a ETD14 core. This means it can transfer around 1.2 kW before it is thermally overloaded.

The amplifier modules are ZAPpulse 2.3SE's. That's why i call it Lightforce 1, i plan to upgrade it to Lightforce 2, when new modules are available.

More to come....
Click to expand...


Hi Lars
Just wondering how the SMPS was coming along. Have you made any changes, improvements, etc ... :cool:
 
L

Lars Clausen

Member
Joined 2003
newclassd.com
KLe i have been testing out some different gate drivers, but so far not found the right solution. I was playing with something that could also maybe be used for a new high speed class D amplifier in the future. But so far only partly success.

Also i have been working on other projects of course ;-)

I will come back as soon i have something that works 100%

All the best from Lars
 
P

phase_accurate

Member
Joined 2002
Maybe we are not talking about the same thing. I am talking of something that is 10 times faster than the fastest opto-couplers and that has ten times the transient immunity of an IR2110 (30 kV/us). There are even two-way ones available that could be used for adaptive dead.time control.

http://www.nve.com/Downloads/il710.pdf

For the PSU gate-drive xformers might pronbably work quite well BTW.

Regards

Charles
 
KLe

KLe

Banned
Joined 2005
Lars Clausen said:
KLe i have been testing out some different gate drivers, but so far not found the right solution. I was playing with something that could also maybe be used for a new high speed class D amplifier in the future. But so far only partly success.

Also i have been working on other projects of course ;-)

I will come back as soon i have something that works 100%

All the best from Lars
Click to expand...

Hi Lars
Thankyou for the update ... keep us posted ... :D

Lars, what are your thoughts on what Charles has just posted? :whazzat: :cannotbe: :bigeyes:
"I am talking of something that is 10 times faster than the fastest opto-couplers and that has ten times the transient immunity of an IR2110 (30 kV/us). There are even two-way ones available that could be used for adaptive dead.time control."
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom