Making car amplifier SMPS with tl494 + DC Protection

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

I dont have time to draw schematics, but here are my pbc, bottom + top(component) + bottom&componentlayer.

I havent tested thiss fully. So I dont suggest you to build it as final solution . Most of shematics is from valveaudio with minor modifications .
All inductors are about 100uH. (4 inductors)

Mosfets : irfz44N (buz11 also worked).
Gate resistors are 4.6 ohm (vertical)
External wires are drawn with red and green color.

In this smps I used EE core from PC power supply. ETD49 cores will wait for final solution :)

Like I said when I find time (this week or week after), I'll do full test with scope .

Bye
 

Attachments

  • smps11-pcb.zip
    95.8 KB · Views: 4,025
coupling cap

originally posted by Tensop
a coupler cap between 12v input and the centretap on the primary side, care to explain in more detail?

Memory again. The coupling cap is used for the half bridge, not push-pull topology as far as I know. I should be careful when relying on memory. My apologies. The rest of the post should be correct. To protect a push-pull converter from saturation due to flux "walking", a current mode controller should be used, or a voltage mode controller with pulse by pulse current limiting. Best regards.
 
Grucho said:
Hi.

I dont have time to draw schematics, but here are my pbc, bottom + top(component) + bottom&componentlayer.

I havent tested thiss fully. So I dont suggest you to build it as final solution . Most of shematics is from valveaudio with minor modifications .
All inductors are about 100uH. (4 inductors)

Mosfets : irfz44N (buz11 also worked).
Gate resistors are 4.6 ohm (vertical)
External wires are drawn with red and green color.

In this smps I used EE core from PC power supply. ETD49 cores will wait for final solution :)

Like I said when I find time (this week or week after), I'll do full test with scope .

Bye
hi grucho
Nice pcb design. Hey how to design menualy or by autoruting with wich software and version. How to convert pcb file to pdf or gif have you any converter?
 
Hi.

I used Eagle for PCB and manual routing (autoroute gives me labyrinth pcb) .
I exported from Eagle to PS.
PS -> PDF with AdobeAcrobat
PDF+PDF -> GIF with AdobePhotoshop


Anyways I have done some small test with scope, but without regulation (I puled out 4n35). There is no big spikes, 1v maby without load, and practicly no spikes with small load.

But 5 Watt resistor (56 ohm) gets very hot with load .

When I get output capacitors I'll do test with pwm (4n35).

bye
 
Hello

Hello

Fist of all I have seen that car amplifier smps is very interesting project. The purpose is to make the smps work efficiently and to make PCB (not very large).
I have put my smps schematic + PCB and car amplifier PCB. These designs I have tested and they work very well. They work like original car amplifier buy from store, like my Kenwood kac-719.
The critical problems for making car amplifier are PCB and winding ferrite core. The PCB who I posted is very small and easy to build and very efficiently.
For winding the toroidal ferrite core uses calculating from www.valveaudio.tripod.com ,I have use them and from first design I have complete smps.
For winding EE ferrite core uses same calculating with difference first wind primary then secondary.
Just read the full design for making car amplifier from valveaudio site couple of times, and use some designs PCB of his site who is open with protel 99se and you will complete the smps.
 
Hi.

Well I tried my smps 3 weeks ago with some mosfet amplifier, and I played very good (looud), but when I looked some stuff with scope I found some big voltage spikes. Spikes have reduced when I used 4 more tansistors to drive mosfets instead using only sg3525.
Using 5watt 56ohm resistors+100nF cap reduced spikes to minimum value , but they heat alot.

Also I think that I found some problem with opto feedback when smps is working in pwm.

I dont want to have time bomb in my car. If smps will not work perfectly I will not put it in my car.

Greetings.
 
Nice PCB, gave me a few ideas on my design, here is my pcb based on immo_g's design + extras its a few versions out of date.. its a bad idea to be doing work on a pcb design at both tafe and home at the same time :D :D things get out of sync.
the version at home as optical feedback, remote switch on and different sized pads on the traffo+ general work
An externally hosted image should be here but it was not working when we last tested it.


i did a version one etch of this, and finished it up today(excl traffo and output) and theres a very nice clean switching waveform :D however on start up the waveform being produced on the CRO is very erratic
 
Hi.

Well isnt 3 mosfets too much for sg3525 ??

Anyway in my PCB this 2k2 resistors that goes from base of BD transistors to GND must be removed, I get cleaner square wave without them .

btw: Tensop I dont understand word CRO in sentence "waveform being produced on the CRO is very erratic" . Also when you test your smps can you report does these E$34 resistor heats, and what resistance it was ?

bye
 
Commentable Thoughts

Hi Guys

I have made an amp + smps for my Contessa Classic car.

The smps uses
SG3525 PWM IC
12 X IRFZ44N mosfets
12 X 47ohm Gate resistors
1 X E-I CRGO 4 No. Iron Core Transformer :primary=12-0-12 Multifillar Winding and Secondary 50-0-50 Windings.
one Vero board
2 X 100mFd 25V Caps
2 X 0.1mfd 50V ceramics cap
Couple of other resistors and caps etc.
Frequency is 250Hz
No - Feedback is applied , therefore efficiency is high.
1 X 3.5mFd/250V cap is directly connected to secondary of transformer to suppress spikes.
2 X small heat sinks.

Now the results
No heating on full load only simple warming of sinks
650W power capability
Still alives and is breathing in my car

IF any one interseted mail me.
 
Hi,

Well I think I have problems with PWM. When I use non-regulated supply everything is OK, no spikes or anything. But when I switch to PWM I get strange readings on scope on primary and secondary. Gates looks at it should at any duty cycle, but look at primary. I dont know is this ok ?? this is about 40% duty cycle. After rectification everything is normaly I get targeted voltage.
 

Attachments

  • scope trace.gif
    scope trace.gif
    18.9 KB · Views: 4,858
Hi amp man

Amazing results, constant output power, no overheat, now you got to help us here, I don't know the dimensions of your E-I iron, from where did you get it anyway.

What wire gauge did you use,and why did you use the 3.5mF capacitor on the secondary directly,and how did you wind your transformer.

Now if I bring an old 5 Amp 48 volt ordinary mains transformer, can I use it, if so, are the number of windings still the same, and what wire gauge should I use,and how many total mosfets should I use, I mean how much power can I obtain from it.

BTW, how should I wind the transformer,there is a plastic core for the primary,and another one for the secondary,should I wind the primary in clockwise and the secondary in counter-clockwise for example,or both of them in clockwise.

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