HELP WITH DC-DC BOOSTER CONVERTER

120 views
Skip to first unread message

MrThe50sanchez

unread,
Jun 23, 2020, 4:22:19 PM6/23/20
to neonixie-l

Hello Again, hope everyone is good!

I was trying to make a DC-DC booster converter using the LT1308B/A device, I made a PCB for it, doing the same as the datasheet of this device says, but  I don´t get the High Votage, and that`s
why im here asking for help, I tried a lot of things put don`t know what it can be.

This is the schematic I use  (This one is from the datasheet) :


The PCB I made is the same as this schematic and using the same components, Im using a rechargable battery CR2 800mAh 3.0V Li-on.

Any Ideas of what I can do to find what is going wrong?

*I have LBI and LB0 from the LT1308 Device not connected to anything.
*/SHDN pin is connected to Vin, Positive Voltage to It works.
*Because I want 180V output voltage I have resistor from FB R1 = 47K and R2 = 300 Ohm to get 1,21V Feedback.
*The capacitors Im using (10nF 250v) are not electrolitic, they are normal ones.


Thankss Guys!

gregebert

unread,
Jun 23, 2020, 5:26:00 PM6/23/20
to neonixie-l
First thing is to get a scope and start looking around.

1. You should see the SW pin pulsating. If not, what's happening on the SHDN and FB pins ?
2. What is your battery voltage reading when connected and disconnected ? Are you sure it's 3.0V ? All of the Li-ion batteries I've used are "3.7V", which really means 3.6 to 4.2V depending upon the sate-of-charge and load-current.

The VC and FB pins are sensitive to noise. Make sure LBI is floating.

Kevin A.

unread,
Jun 23, 2020, 5:40:06 PM6/23/20
to neonixie-l
What does your layout look like? Show us a picture of the pcb with the layers visible. 

Just the layout of the board can make or break a DC-DC converter with all of the high current transients and control signals.

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/7d2fbb96-d88c-4341-b89a-2ab723f403dao%40googlegroups.com.

MrThe50sanchez

unread,
Jun 24, 2020, 6:45:13 AM6/24/20
to neonixie-l
Hello, Thanks for your answer.

·Is sad but I dont have and Scope to watch what is going on with the SW pin from te LT1308 device.

·My battery voltage when I connect it is 3.0V and the readings in the Vin and SW are 3.0V. I add you a pic from the battery Im using and also the schematic of the PCB.
Is the LB1 and LB0 floating like that? Im from Spain and have searched and Understand that floating is leaving it not connect to anything, just like that.

MrThe50sanchez

unread,
Jun 24, 2020, 6:46:01 AM6/24/20
to neonixie-l
Hello! Thanks for the response!

I add here again the PCB layers that Im using.


El martes, 23 de junio de 2020, 23:40:06 (UTC+2), Kevin A. escribió:
What does your layout look like? Show us a picture of the pcb with the layers visible. 

Just the layout of the board can make or break a DC-DC converter with all of the high current transients and control signals.

On Tue, Jun 23, 2020, 4:22 PM MrThe50sanchez <the50s...@gmail.com> wrote:

Hello Again, hope everyone is good!

I was trying to make a DC-DC booster converter using the LT1308B/A device, I made a PCB for it, doing the same as the datasheet of this device says, but  I don´t get the High Votage, and that`s
why im here asking for help, I tried a lot of things put don`t know what it can be.

This is the schematic I use  (This one is from the datasheet) :


The PCB I made is the same as this schematic and using the same components, Im using a rechargable battery CR2 800mAh 3.0V Li-on.

Any Ideas of what I can do to find what is going wrong?

*I have LBI and LB0 from the LT1308 Device not connected to anything.
*/SHDN pin is connected to Vin, Positive Voltage to It works.
*Because I want 180V output voltage I have resistor from FB R1 = 47K and R2 = 300 Ohm to get 1,21V Feedback.
*The capacitors Im using (10nF 250v) are not electrolitic, they are normal ones.


Thankss Guys!

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neoni...@googlegroups.com.

Ian Vine

unread,
Jun 24, 2020, 8:43:35 AM6/24/20
to neonixie-l
"I have resistor from FB R1 = 47K and R2 = 300 Ohm to get 1,21V Feedback."

The FB resistors in the datasheet are 10M and 34.8K. Just thinking you might want to use similar values. Rough calculation your FB divider will draw more current than the device can supply. I might be wrong but worth looking at.

Cheers
Ian

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

To view this discussion on the web, visit

Paul Andrews

unread,
Jun 24, 2020, 8:49:49 AM6/24/20
to neonixie-l
Read the layout recommendations in the LT1308A datasheet. You will need to adapt it to this configuration.

What is the voltage rating of the 47k resistor? What are the voltage ratings of the 10nF capacitors? What is the voltage rating of BAV21W? What is the saturation current of the transformer?

MrThe50sanchez

unread,
Jun 24, 2020, 9:03:52 AM6/24/20
to neonixie-l
Hello!

Yes!  I would try that, the thing is that if I put those values, Ill get 350V 1,2mA as the datasheet says.

That´s why I made this calculations :

If I want to get a Vout of 180V for supply a Nixie tube, I need to put a R1 =47K and R2 = 300. that`s Why I putted those!

thanks a lot!
To unsubscribe from this group and stop receiving emails from it, send an email to neoni...@googlegroups.com.

To view this discussion on the web, visit

Kevin A.

unread,
Jun 24, 2020, 9:04:43 AM6/24/20
to neonixie-l
So if you really want to make this project work, then it's probably best you start with a fresh layout. The PCB layout that you started with needs a lot of work.

Believe me, I made the same mistakes when I first started designing high voltage power supplies. Take a look this article I wrote which describes how I learned about and designed a high voltage boost converter: https://neonkev.com/2019/08/18/project-20-watt-high-voltage-boost-converter/

Your design is a flyback converter, but a lot of the same pcb design still applies. I'll also find some of the articles I read which really helped with my design. 

Once we have a good layout, then we can work on picking the right values for the feedback resistors and choosing all the other components. There is a bit of a learning curve but be patient, read carefully and learn, and you will make a working design soon enough! 

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/88551fe2-e03e-4d81-aaf9-794b0a7f5f68o%40googlegroups.com.

MrThe50sanchez

unread,
Jun 24, 2020, 9:06:07 AM6/24/20
to neonixie-l
Hello !

Yes, I used the components that the Datasheet says, I got it from here:


The thing is, that a Triple Voltage Circuit uses electrolitic capacitors, this is not the same configuration as a voltage triple circuit, but, Can It be that the capacitors I added are not electrolitic? I just
putted normal ones (SMD 10nF 250V).

Thanks a Lot!

Paul Andrews

unread,
Jun 24, 2020, 9:08:46 AM6/24/20
to neonixie-l
Where is D4 in your circuit? It should be between the transformer pin connected to SW and the transformer pin connected to ground. 

Paul Andrews

unread,
Jun 24, 2020, 9:13:12 AM6/24/20
to neonixie-l
Also, if the datasheet shows transformer pin 6 connected to ground and pin 4 connected to the output. You have these swapped.

Ian Vine

unread,
Jun 24, 2020, 9:28:38 AM6/24/20
to neonixie-l
"
Yes!  I would try that, the thing is that if I put those values, Ill get 350V 1,2mA as the datasheet says.

That´s why I made this calculations :

If I want to get a Vout of 180V for supply a Nixie tube, I need to put a R1 =47K and R2 = 300. that`s Why I putted those!
"

You need to look at the current in the FB loop. The datasheet values give a feedback current around 35uA, so for 170v  and FB i of 35uA, you're looking at something around 5M for R1

Cheers
Ian

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

To view this discussion on the web, visit

Kevin A.

unread,
Jun 24, 2020, 12:10:12 PM6/24/20
to neonixie-l
sanchez,

have a look at these articles by rohm semiconductor on DC-DC converter layout: https://techweb.rohm.com/knowledge/dcdc/dcdc_pwm/dcdc_pwm03/2734

this is quite a handy guide on how to optimize the layout of your board.

since you are also using an LT part, I would also highly recommend using LTspice (free design and simulation software from LT) to model the LT1308A on your computer and do simulations with different resistors, capacitors, transformers, etc...


if you use these design tools you can save yourself a lot of troubleshooting and frustration 

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/0c171ec5-8a36-4e9b-9d57-6fecd1d37101o%40googlegroups.com.

Kevin A.

unread,
Jun 24, 2020, 12:15:08 PM6/24/20
to neonixie-l
I've also been designing some nixie flyback power supplies lately. I will do a full writeup on my website of the project when the project is complete. 

I recommend these tools because I have experience with them and they work! 

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
flyback_psu_2D_1.PNG
flyback_psu_3D_1.PNG

Kevin A.

unread,
Jun 24, 2020, 12:39:30 PM6/24/20
to neonixie-l
lets start with problems with the low voltage part of your board. your input power and ground are running through long, skinny traces. this is not good... you want to minimize the L and R in these parts of the board. 

Start with making the bottom layer of the board one big ground plane. Just make sure to add a keepout zone below the transformer. See the example in the picture below from my design. This should make sure every circuit that is connect to ground has a low impedance path which will help minimize ground problems.

Next I would suggest paying special attention to your power traces. Especially the low voltage/high current path. Try to "visualize" where the current is going, and how much current is actually flowing through different parts of the board. I tried to illustrate the low voltage currents on the top copper layer in yellow, and the high voltage in blue. 

Remember than in a switching power supply, these currents are being turned on and off rapidly. The strong magnetic fields associated with this can interfere with your feedback circuitry and other small signals on the board. Try to keep the small signals away from the transformer and the large power traces. 

And remember... the copper is essentially "free". Circuit boards typically made by etching copper away... so you're paying for it any which way. Use as much of it as possible! 

Lets see what you come up with. 
psu_annotation_1.jpg
flyback_psu_2D_bott_copper_1.PNG
flyback_psu_2D_top_copper_current_paths_power.png

Ian Vine

unread,
Jun 24, 2020, 1:25:05 PM6/24/20
to neonixie-l
Nice. Be good to see these

Cheers
Ian

MrThe50sanchez .

unread,
Jun 24, 2020, 2:17:45 PM6/24/20
to neoni...@googlegroups.com
Thanks a Lot!

the main reason of this is learning always, love electronics, so I will take a really good look to the link you attached.

Let´s see If I find a way to change the design to it works.

Thanks a lot I will tell you How is going on


MrThe50sanchez .

unread,
Jun 24, 2020, 2:18:50 PM6/24/20
to neoni...@googlegroups.com
Oh F**CK yes,

I made that mistake because I putted in the wrong direction the transformer, That`s a silly mistake.

Thanks!

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/f60d6ff7-a6d6-42f6-bac0-249c239f0340o%40googlegroups.com.

MrThe50sanchez .

unread,
Jun 24, 2020, 2:22:13 PM6/24/20
to neoni...@googlegroups.com
This is Crazy, I want to learn how to build this things.

Let's read a lot!!


MrThe50sanchez .

unread,
Jun 24, 2020, 2:32:09 PM6/24/20
to neoni...@googlegroups.com
I really appreciate your time for helping me with this,
 
Going to start making change's, learning, and I will tell you I im doing.

Thanks a lot 🤘

Reply all
Reply to author
Forward
0 new messages