Ta3020 linear psu Design help

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi guys.
I am trying to design a liner power supply for my new TA3020(v3b by cristi), but i need help.

From the datasheet of the TA3020 v3b i got that
1_untitled.JPG
So, a 36-0-36 500VA should be enough.

My ideas is to start from the aleph 5 power supply, and then adapt it for my case.
An externally hosted image should be here but it was not working when we last tested it.


Let's start the rework
  • Graetz bridge: 35A 400V VISHAY
  • Caps: 3 10.000uf 63v RIFA electrolytic caps for branch
  • According to this thread no bypassing MKP capacitors.

Here is my questions
  • How many bridge Should i use, only one or two?
  • Should I remove the 2 inductance?
  • Are my ideas correct according to the work spec?

At the moment I have not other questions in my mind, but I know that there will be more.


PS: Don't hate me for my ENG. I am trying my best.
Thank you
 
If you choose to use a linear supply instead a smps, first i would recommend to use a toroidal transformer with a center tap secondary, at half of the output AC voltage. for example, a 230VAC to 2x40V AC transformer can be used, which will provide after rectification and filtering about +-55V DC at no-load. depending on the transformer power rating, the voltage could drop even to +-50V and below at high current demand.

I would recommend to use a single rectifier bridge, rated at least 15A 200V, but if you can get a 25A or 35A is even better. the center tap of the transformer should be the power Ground, and each AC winding end should be connected to an AC terminal of the bride. the advantage of using a single rectifier instead of two, is that the voltage drop on th diodes will be lower, the current have to pass through a single diode at towards each supply rail instead of two, as for two rectifier bridges version.
Also note that the working voltage or reverse voltage should not be higher than 600V, otherwise you will experience increased voltage drop on the diodes.
You can use the schematic from here http://www.diyaudio.com/forums/connexelectronic/158603-linear-power-supplies.html#post2208810 to build a simple and robust linear power supply. don't forget about the bleeder resistors and eventually you can add the led's too.
 
Last edited:
If you choose to use a linear supply instead a smps, first i would recommend to use a toroidal transformer with a center tap secondary, at half of the output AC voltage. for example, a 230VAC to 2x40V AC transformer can be used, which will provide after rectification and filtering about +-55V DC at no-load. depending on the transformer power rating, the voltage could drop even to +-50V and below at high current demand.

I would recommend to use a single rectifier bridge, rated at least 15A 200V, but if you can get a 25A or 35A is even better. the center tap of the transformer should be the power Ground, and each AC winding end should be connected to an AC terminal of the bride. the advantage of using a single rectifier instead of two, is that the voltage drop on th diodes will be lower, the current have to pass through a single diode at towards each supply rail instead of two, as for two rectifier bridges version.
Also note that the working voltage or reverse voltage should not be higher than 600V, otherwise you will experience increased voltage drop on the diodes.
You can use the schematic from here http://www.diyaudio.com/forums/connexelectronic/158603-linear-power-supplies.html#post2208810 to build a simple and robust linear power supply. don't forget about the bleeder resistors and eventually you can add the led's too.
Thank you.

A little question.
175060d1275843492-linear-power-supplies-ps10k80.jpg


Are C1-C2-C7-C8-C6-C12 mkp caps?
 
Your transformer is a bit light, but should be OK unless you are planning PA service. Bigger transformers will have better regulation, although I doubt you'll notice the difference between 5% and 10%

The specified bridge will work.

36 VAC will give you around 49V after rectification. You can go up to 40 VAC if you haven't already bought the transformer.

If your transformer is really 36-0-36 rather than 2 separate 36 VAC windings, you MUST use a single bridge. If two separate windings you can either tie the windings in series to create a center tapped winding and use a single bridge or rectify them individually.

You don't need the inductors in your application.

You probably won't have an issue without MKP bypass. You might try with and without. IF there is HF noise from the Tripath on the rails your electolytic caps may present a high impedance and the MKPs may be needed. OTOH, some have reported that the transformer inductance and the MKPcan cause resonances. You'll have to see what suits your application.

The 100nf and 220 nf could be MKP or MKT. Not strictly needed.

If you choose to use the LED power indicator, you will need to increase the resistor in series with the LED. You might want to go from one rail to ground to reduce the power wasted. Choose an LED current to suit your brightness requirements, say 5-10mA. Subtract the LED voltage drop from your rail voltage (1.2-5V depending on color) divide the remainder by the chosen current to get the resistance required. Multiply the voltage across the resistor by the current then by 2 for a safety factor to get the power rating.

Also, R1 and R2 should be higher - 3-4K 3W.

Don't use MUR860s for this application. Use the bridge you specified or MUR2060 to give you a rating margin.
 
Last edited:
Your transformer is a bit light, but should be OK unless you are planning PA service. Bigger transformers will have better regulation, although I doubt you'll notice the difference between 5% and 10%

The specified bridge will work.

36 VAC will give you around 49V after rectification. You can go up to 40 VAC if you haven't already bought the transformer.

If your transformer is really 36-0-36 rather than 2 separate 36 VAC windings, you MUST use a single bridge. If two separate windings you can either tie the windings in series to create a center tapped winding and use a single bridge or rectify them individually.

You don't need the inductors in your application.

You probably won't have an issue without MKP bypass. You might try with and without. IF there is HF noise from the Tripath on the rails your electolytic caps may present a high impedance and the MKPs may be needed. OTOH, some have reported that the transformer inductance and the MKPcan cause resonances. You'll have to see what suits your application.

The 100nf and 220 nf could be MKP or MKT. Not strictly needed.

If you choose to use the LED power indicator, you will need to increase the resistor in series with the LED. You might want to go from one rail to ground to reduce the power wasted. Choose an LED current to suit your brightness requirements, say 5-10mA. Subtract the LED voltage drop from your rail voltage (1.2-5V depending on color) divide the remainder by the chosen current to get the resistance required. Multiply the voltage across the resistor by the current then by 2 for a safety factor to get the power rating.

Also, R1 and R2 should be higher - 3-4K 3W.

Don't use MUR860s for this application. Use the bridge you specified or MUR2060 to give you a rating margin.

I still have to buy everything so I can do every change in freedom.
Let's take Cristi's scheme as ex
  • C1-C2-C7-C8 100nf, 400/250 DC/AC voltage is enough?
  • C6-C12 220nf, 400V
  • Should r1 and r2 be non inductive?

What's about this trafo?
http://ae.rsdelivers.com/product/nu...idal-transformer-800va-2x40v-o-p/2241590.aspx
 
Last edited:
I still have to buy everything so I can do every change in freedom.
Let's take Cristi's scheme as ex
  • C1-C2-C7-C8 100nf, 400/250 DC/AC voltage is enough?
  • C6-C12 220nf, 400V
  • Should r1 and r2 be non inductive?

What's about this trafo?
0800P1-2-040 Nuvotem | Toroidal transformer,800VA 2x40V o/p | 224-1590 | Welcome to RS Online

I got the trafo.
2x40V 800VA.

About r1-r2, 3-4Kohm?

Should i also make 2 filter like those



The first (1), a smaller value cap of 100uF in parallel with a 100nF capacitor filter out any RF signals or other disturbances , and finally (2), an RC network, 1 Ohm/17W in series with 680nF, gets rid of any residual capacitor inductance.
 
R1 and R2 in the post 4 schematic should be 3-4K, 3-5W

You don't need the additional filters, but if you want to use them any caps rated over 63V will be fine.

The resistors in the snubber (filter #2) don't need to be 17W, 1W would be fine. The proper values of the cap and snubber depend in large part on the transformer. Search the threads on snubber design, its fairly complex. You won't do any damage, but you aren't likely to do a heck of a lot of good with random values.
 
R1 and R2 in the post 4 schematic should be 3-4K, 3-5W

You don't need the additional filters, but if you want to use them any caps rated over 63V will be fine.

The resistors in the snubber (filter #2) don't need to be 17W, 1W would be fine. The proper values of the cap and snubber depend in large part on the transformer. Search the threads on snubber design, its fairly complex. You won't do any damage, but you aren't likely to do a heck of a lot of good with random values.

I have read something. Sounds like snubber circuits are usually used on SMPS.
So I shouldn't need it. But I will try the best solution.

Are the bridges connect correctly as show in this schematics?



Hope this is the last question :p

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