Switched Mode Power Supplies (SMPS) - Page 48 - diyAudio
Go Back   Home > Forums > Commercial Sector > Vendor Forums > Inactive Vendor Forums > Connexelectronic

Connexelectronic Commercial Vendors large & small hawking their wares

 
 
Thread Tools Search this Thread
Old 4th March 2012, 03:35 PM   #471
diyAudio Member
 
Esperado's Avatar
 
Join Date: Jul 2011
Location: France
Hi, Cristi,
I would appreciate some answers to my questions,
Switched Mode Power Supplies (SMPS)
as i need them to send my order.
__________________
Ultimate Protection and more.The Only Source of Knowledge is experience, everything else is just information” ©A. Einstein
 
Old 5th March 2012, 08:50 AM   #472
Cristi is offline Cristi  Europe
diyAudio Member
 
Join Date: Dec 2008
Location: Dortmund+Shenzhen
Jaques, i sent you the boards already, i will send you a mail with the tracking info.
Christophe, i will try to answer to your questions:

1. One SMPS800R is enough to supply one 250W class D amplifier module and one ~ 150W class AB amplifier. you can consider one SMPS500R but this might be at limit, as the class AB module draws more current than an equivalent power class D amplifier, roughly twice.
2. Separate SMPS might be a better choice, as the interractions between amps on the supply path are eliminated. just consider that the class D amplifier will experience pumping effect, since you will use it for low frequencies mostly. if the class AB amplifier does not have high PSRR it's performaces might be affected.
3. Adding capacitors is recommended if you want to reduce bus pumping or Hi-freq noise, if you also add some inductors in series with each supply rail. the total capacitance of the caps should not be double than the existing value of the caps from the smps or the smps might not start (enter in overcurrent protection mode during start-up, due to large inrush current)
4. The switching frequency of the resonant converters cannot be synchronized, as depends on the load, mains voltage and resonant tank characteristics. And is not even required. the switching frequency of the PWM power supplies or fixed frequency quasi-resonant type smps can be synchronized, but with additional hardware.
__________________
www.connexelectronic.com complete assembled amplifier and SMPS
 
Old 5th March 2012, 01:23 PM   #473
diyAudio Member
 
Esperado's Avatar
 
Join Date: Jul 2011
Location: France
Quote:
Originally Posted by Cristi View Post
2. Separate SMPS might be a better choice, as the interractions between amps on the supply path are eliminated. just consider that the class D amplifier will experience pumping effect, since you will use it for low frequencies mostly. if the class AB amplifier does not have high PSRR it's performaces might be affected..
It is the case: My class AB amp is a current feedback one (I.E. lowPSSR) with a 3Mhz power bandwitch.
www.esperado.fr - Le crescendo revisité
But i always prefer using a single Supply for harmonic dynamic coherency.between bass and treble during hight kick drums transients.
In the same spirit, i always prefer single power supply for the two channels for space image stability instead of separate ones. But, in my case, it is not possible (amps in the enclosures).
Quote:
Originally Posted by Cristi View Post
3. Adding capacitors is recommended if you want to reduce bus pumping or Hi-freq noise, if you also add some inductors in series with each supply rail. the total capacitance of the caps should not be double than the existing value of the caps from the smps or the smps might not start (enter in overcurrent protection mode during start-up, due to large inrush current)
Well, using some air inductors, what would be the best values for the inductance ? What do you think about an other solution: adding a cap multiplier transistor and caps on each amps rails ?
www.esperado.fr - Alimentation Crescendo
I will lose some volts, but increase PSSR and HF filtering ?
And a last question. if i add a power resistance in serial in each rail during the starting time to reduce the inrush charging current, i suppose i can increase the total value of power capacitances with no issue for your SMPS ?
My system would be 3 SMPS800R.
For the sub, one SMPS, followed by a self and additional caps.
For the big satellites, one SMPS, each rail followed by a self and 2 cap multipliers (one for each amp) and several paralleled additional capacitances .
I cannot afford any hf noise, as my speakers are hight efficiency 98db/w units.
Did-you agree ?
I will publish an article of my final system on my site including your SMPS if it works great, with your permission.
__________________
Ultimate Protection and more.The Only Source of Knowledge is experience, everything else is just information” ©A. Einstein

Last edited by Esperado; 5th March 2012 at 01:34 PM.
 
Old 7th March 2012, 02:20 AM   #474
Cristi is offline Cristi  Europe
diyAudio Member
 
Join Date: Dec 2008
Location: Dortmund+Shenzhen
Chistophe, Please wait few days, I just made a new SMPS300R version, which has LC filters on each output. On the actual version, there are 4 electrolytic caps on the secondary side, for each rail there are two caps connected in parallel, and for the single voltage version all 4 caps are connected in series. On the new SMPS300RE an CLC filter is implemented on each rail, the capacitors are not connected in parallel but an inductor is added on each rail between the caps, reducing the ripple even more.

I am not an adept of using air-core inductors, because they are bulky, and because they radiate a lot, requiring shielding or placing them far away from sensitive circuits. with the modern ferrite cores, can easy get at least the same performance of an air-core inductor or even better.
Adding a capacitance multiplier might be considered as last resort, if excesively low ripple is required, but take into account that there is power loss on the transistors, and complexity of short-circuit protection for that transistors. the overall efficiency will decrease. I find it odd using a high-efficiency power supply as a front end for a lossy circuit.
Adding a resistor in series is a bad idea, the output voltage regulation is lost in this way. you can add as much capacitance as you want (use good-sense limits) if the soft-start time is long enough to allow capacitors to charge with a limited current and without the smps entering in overcurrent protection mode.

I believe that the most important aspects to consider will be the correct placement of the boards into housing and correct wiring, a GND loop, wrong wiring, signal wires close to the SMPS transformers will harm more than you can save with expensive extra-circuits. Careful planning is required, and perhaps few wiring arrangements should be considered.
__________________
www.connexelectronic.com complete assembled amplifier and SMPS
 
Old 7th March 2012, 03:15 AM   #475
diyAudio Member
 
Esperado's Avatar
 
Join Date: Jul 2011
Location: France
Quote:
Originally Posted by Cristi View Post
Adding a resistor in series is a bad idea, the output voltage regulation is lost in this way
It it just during the charging time of the additional secondary stage caps (with a delay). They will be short-circuited with a relay when caps arrive at 80/90% of the final V. That way, i can use big capacitance. But, if your SMPS is able to limit itself the current with no issue during long soft start (5sec), it will not be needed.

I don't care to lose 5v on 75V with a capacitance multiplier (don't forget your SMPS is still regulated, not so much more margin than the 0.7V+ ripple needed.). I will need-it with my AB amp witch need perfect power with no hf noise. And, if i care for weight (they will be in the enclosures) and price, i don't care too much about efficiency.
It will isolate the two amps from each other. I don't care neither for protection: a fuse is enough and i protect the amp outputs in a very secure and fast way.

I will wait for your new smps if they are available soon and able to give-me the + & - 75v i need.
__________________
Ultimate Protection and more.The Only Source of Knowledge is experience, everything else is just information” ©A. Einstein
 
Old 7th March 2012, 05:24 PM   #476
avr300 is offline avr300  Denmark
diyAudio Member
 
avr300's Avatar
 
Join Date: Sep 2005
Location: DK
Quote:
Originally Posted by Cristi View Post
Chistophe, Please wait few days, I just made a new SMPS300R version, which has LC filters on each output. On the actual version, there are 4 electrolytic caps on the secondary side, for each rail there are two caps connected in parallel, and for the single voltage version all 4 caps are connected in series. On the new SMPS300RE an CLC filter is implemented on each rail, the capacitors are not connected in parallel but an inductor is added on each rail between the caps, reducing the ripple even more.
If this modification is to enter the SMPS180 arena - I could have waited ordering my 180'.

 
Old 8th March 2012, 03:04 AM   #477
flyboi is offline flyboi  Malaysia
diyAudio Member
 
Join Date: Dec 2004
Location: .
Send a message via ICQ to flyboi
Quote:
Originally Posted by Cristi View Post
Chistophe, Please wait few days, I just made a new SMPS300R version, which has LC filters on each output. On the actual version, there are 4 electrolytic caps on the secondary side, for each rail there are two caps connected in parallel, and for the single voltage version all 4 caps are connected in series. On the new SMPS300RE an CLC filter is implemented on each rail, the capacitors are not connected in parallel but an inductor is added on each rail between the caps, reducing the ripple even more.

Hi Cristi!

Interested in this version. How low is the ripple value(millivolt/microvolt) with this CLC filter in place?


Eddy.
 
Old 8th March 2012, 12:51 PM   #478
Cristi is offline Cristi  Europe
diyAudio Member
 
Join Date: Dec 2008
Location: Dortmund+Shenzhen
Christophe, i got it now, i haven't read carefully before. I thought that the resistors will be connected permanently. The soft-start time can be theoretically as long as you need but some modification is required, to allow the bias winding to supply voltage for the primary side section during start-up when the sw. freq is far from resonant freq. which allows secondary current limit. I have to try to be sure it can be done on the existing boards. with the current configuration, if very large capacitors are added, the smps will start, try to ramp-up the voltage in the preset start-up time, and if the current draw will exceed the maximum allowed the smps will enter in overcurrent protection mode, then after 1-2 seconds will try to restart. if the caps were not discharged in this interval, on the second-third try might start-up. overall it would take few seconds as well. but the current stress on the smps components is much higher, and could affect the proper operation and reliability if at each start-up will cycle few times through OC protection mode.

A 5V voltage drop from a 75V rail is indeed small, but heatsinks are required for the series regulators, and more important, an over-current circuit protection with fold-back characteristic. at that voltage, to limit the current at few amps would burn any TO220 transistor in less than one second. also, you have to make sure that the Vce of the transistor is higher than the smps output voltage. Although the SMPS300R has no heatsink, when operate near it's maximum output power or if supply class AB amplifiers, is recommended to be equipped with a heatsink, which can be fitted onto the case wall or heatsink. The board allows such heatsink to be mounted, and even more power can be obtained if the heatsink is used.

Jesper, If you want, i will check with the shipping company tomorrow what's the status of your package, if hasn't left china yet and can be replaced, i can change the order to another power supply. please contact me asap to tell me if you want.

Eddy, I haven't measured yet on the batch boards, but on the prototype and simulation was around 7-11 mV pk-pk for a +-45V version at 2A. the inductors and capacitors used have a great influence on the ripple value. Later, i will try to make some measurements using spectrum analizer to see all the noise spectrum at different current draws.
__________________
www.connexelectronic.com complete assembled amplifier and SMPS
 
Old 8th March 2012, 02:54 PM   #479
diyAudio Member
 
Esperado's Avatar
 
Join Date: Jul 2011
Location: France
Thanks, Cristi, for all the time you gave. Do not worry, i was an electronic engineer during the 1970 decade in the research and development office of the most important hifi manufacturer at this time.
My questions where just about YOUR smps, as i'm not experienced with this technology,(i was dedicated in linear PSUs since now).
Everything is clear, now, with your nice answers.
I thing i will not wait for your new model, as i will work on my own filtering. So i will pass my order on your site this night or tomorrow.
About air inductances, (i have plenty of them) if i use them, i do not expect any issue, as they will be *after* your SMPS, so they will only have to deal with the very little remaining hf ripple. No saturation and low impedance (big wires diameter). And i know how to wire, shield (and cool ;-) all that stuff.
Thanks again.
__________________
Ultimate Protection and more.The Only Source of Knowledge is experience, everything else is just information” ©A. Einstein
 
Old 8th March 2012, 04:10 PM   #480
glt is offline glt  United States
diyAudio Member
 
Join Date: Oct 2004
Quote:
Originally Posted by Cristi View Post
...but on the prototype and simulation was around 7-11 mV pk-pk for a +-45V version at 2A. the inductors and capacitors used have a great influence on the ripple value. Later, i will try to make some measurements using spectrum analizer to see all the noise spectrum at different current draws.
Using a "typical" value of 9 mV p2p, we get 4.5mV peak or 3.2mV RMS. That is .007% of 45V

Very close to a typical low voltage regulator spec of 0.003% ripple
__________________
www.hifiduino.wordpress.com
 

 


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
switched-mode power suppies Dimitar13 Everything Else 0 19th February 2008 12:50 AM
PFC Switched-Mode Power Supply N-Channel Power Supplies 57 13th June 2005 07:56 PM
Switched mode power supply MMicke Car Audio 3 19th February 2005 10:42 AM
Switched mode power supplies Magura Parts 13 3rd January 2004 05:44 PM
Switched Mode vs Linear Power Supply royalmp2001 Solid State 4 12th July 2002 08:04 PM


New To Site? Need Help?

All times are GMT. The time now is 01:09 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2