Voltage doubler -Voltage regulator issues

[rocky79]

I am trying to build a voltage doubler using a voltage doubler and a voltage regulator. The whole idea is to get a varying voltage between 4 to 5 v input, double it and regulated it to exactly 5v.
The problem is when i connect the voltage doubler output to the voltage regulator the voltage doubler dies, it stops working meaning it's output will not be double the input anymore, it will be the same as the original input.
I am using the circuit on page 1 of the voltage doubler datasheet.
The Voltage doubler is Ame7700 and the voltage regulator is an ST L78050.
Is it impedance match problem? can i anyone explain what's happening.
Thanks

[oldphart]

this is just a guess,but most switched capacitor voltage translators are quite low current .Are you overloading it? Try a google search for the device,I just tried to ,but all the data sheets are PDF files,and adobe acrobat has screwed up my computor~ must get it fixed one day

[Terry S]

Just had a look at the data sheet and the oldphart is right. Can you post the circuit ?
The AME7700 has a low output power, what did you want to poer with it ?

Terry

[rocky79]

well i am just connecting the output of the voltage doubler to the input of the voltage regulator.It's that simple.The datasheet for the voltage doubler mentions that it's output impedance is 50 ohms.In the datasheet, the output test condition of the voltage doubler uses 10k ohms as a test load resistance.
I measured the input impedance for the voltage regulator and it was in the megaohms.
I think if i added a 10kohm at the output of the voltage doubler to ground,it will solve the problem. I still have to try it..

[Terry S]

Like I asked 'Can you post the circuit ?'
You could test the doubler by checking the voltage output on a meter and adding resistors to load the circuit. This would be a good place to start.

Terry

[INewton]

I'd use a normal inductor switching regulator.

[Mr.RC-CAM]

(1) If you connect just a 9V battery to the input of the 7805 Vreg, does it work correctly?

(2) Did you install the required caps on the 7805, per the data sheet? Without them, the Vreg is prone to oscillate, which will cause problems.

[rocky79]

(1) If you connect just a 9V battery to the input of the 7805 Vreg, does it work correctly?

(2) Did you install the required caps on the 7805, per the data sheet? Without them, the Vreg is prone to oscillate, which will cause problems.
Hello
(1)Yes, it does work nicely
(2) i did, the problem is the voltage doubler, it just stop doubling the voltage after i connect it to the V-regulator. it stop working permanently

[Nicetie]

Hello
(1)Yes, it does work nicely
(2) i did, the problem is the voltage doubler, it just stop doubling the voltage after i connect it to the V-regulator. it stop working permanently

Without going into detail, voltage doubler circuits suck. They aren't good
for anything but reference voltages and VERY low current supply
sources. Use a step-up switching regulator instead. The linear regulator is a current hog too.

Ken K5MBV

[Zlatko]

Have a look at LTC3200-5.

I havent tried it from 5V input but "should" be fine .....
http://www.linear.com/pc/downloadDocument.do?navId=H0,C1,C1003,C1039,C1133, P1888,D2855

If that link doesn't take you directly to the pdf file, go to www.linear.com and search for LTC3200-5, the last page of that pdf shows you a diagram for a regulated 5V output from USB port. Should be able to give you 5V regulated with 3 other capacitors ....

Cheers

[rocky79]

I check it out. I am just looking for a voltage doubler. I might need to get a buck-boost regulator.I need a part that handles input voltage changes from 4.8 to 5.5 v.

That AME7700 truly sucks. I tried to connect a 10k resistor at it's output and it heats up and it's stop working.Just burned the 3rd one :mad:

[rocky79]

That's how the voltage doubler is connected. I used a 10uF for C1 and C2. The datasheet mentions that it can supply up to 100ma. Yet if i attach a 10k the circuit will make a buzzing sound "strange" and it won't work.
Input voltage is = 3.7v.
It only works if it's output is left open.Ultimately it's output needs to be connected to a voltage regulator input.
Any idea what's wrong

[INewton]

That's how the voltage doubler is connected. I used a 10uF for C1 and C2. The datasheet mentions that it can supply up to 100ma. Yet if i attach a 10k the circuit will make a buzzing sound "strange" and it won't work.
Input voltage is = 3.7v.
It only works if it's output is left open.Ultimately it's output needs to be connected to a voltage regulator input.
Any idea what's wrong

Do you have the caps (specifically C1) in the right direction?

[rocky79]

Do you have the caps (specifically C1) in the right direction?
They are ceramic surface mount caps.There is no polarity marks on them.

[Comatose]

http://www.dimensionengineering.com/AnyVoltMini.htm will do exactly what you're looking for, and also can step voltage down if need be (say the input goes to 6v)

If you're looking for a more DIY solution, AnyVolt Mini is a SEPIC topology switching regulator. They're more complicated to design and build than buck or boost regulators, but the information is out there if you want to invest the time. SEPIC is a two inductor, one switch, three capacitor topology. A boost-buck cascade might also get you where you want to go, but that requires two switches.

[Dan Baldwin]

They are ceramic surface mount caps.There is no polarity marks on them.

If they are 10 mfd capacitors they should be electrolytics, and should have polarity markings. Are you sure they are 10 mfd?

Dan

[rocky79]

If they are 10 mfd capacitors they should be electrolytics, and should have polarity markings. Are you sure they are 10 mfd?

Dan

Here is the one i got.587-1352-1-ND (http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?Ref=391740&Row=197612&Site=US) from digikey. It's a ceramic cap

[INewton]

They are ceramic surface mount caps.There is no polarity marks on them.

They can't be 10 uF if they're ceramic surface mount caps. They might only be 10 nF. That poor IC must be getting hot as :censored: trying to supply current with those tiny caps. That's your problem!!!!!!!!!!!!!!!!!

The circuit is too simple not to work. Just go to RS and get a couple of 1 uF to 10 uF aluminum electrolytics to be sure.

[Comatose]

We use 10uf and even 22uf ceramic surface mount caps. They're not cheap, but they exist.

In fact, those ones ARE 10uf. But they're also y5v, which makes them wildly unsuited for this task. Y5V loses almost all of its capacitance(well over 90%) when you put a dc bias onto it. You'll need x5r or x7r grade caps to make the circuit work properly.

Digikey part number PCC2417CT-ND will work much, much, much better and isn't too unreasonably expensive.

[rocky79]

We use 10uf and even 22uf ceramic surface mount caps. They're not cheap, but they exist.

In fact, those ones ARE 10uf. But they're also y5v, which makes them wildly unsuited for this task. Y5V loses almost all of its capacitance(well over 90%) when you put a dc bias onto it. You'll need x5r or x7r grade caps to make the circuit work properly.

Digikey part number PCC2417CT-ND will work much, much, much better and isn't too unreasonably expensive.

Comatose,Thanks for pointing that out to me. That's what i was missing!. At this point i switched back to electrolytic capacitors and it worked fine. But i wanted to regulate the input down to 5 v. So i added a Seiko S-812C50AY-B-G (http://www.mouser.com/index.cfm?&handler=data.listcategory&D=*S812C50AYBG*&terms=S-812C50AY-B-G&Ntt=*S812C50AYBG*&Dk=1&Ns=SField&N=0&crc=true)
5 regulator from mouser.
The problem is at power up the voltage doubler can't double the input voltage. But if i disconnect the voltage regulator from the Vdoubler output and then connect it back on after power up the circuit works. How did you come up with a solution for this?
Thanks

[vintage1]

We use 10uf and even 22uf ceramic surface mount caps. They're not cheap, but they exist.

In fact, those ones ARE 10uf. But they're also y5v, which makes them wildly unsuited for this task. Y5V loses almost all of its capacitance(well over 90%) when you put a dc bias onto it. You'll need x5r or x7r grade caps to make the circuit work properly.

Digikey part number PCC2417CT-ND will work much, much, much better and isn't too unreasonably expensive.

Well strike me pink guvnor..

What a strange capacitor..what use IS a capactor that won't take any voltage across it?

You can't use it for coupling.decoupling or smoothing...only use I can think of is in hugely low frequency filters..but even there, in order to work at any signal level its going to end up with some sort of slowly moving AC (DC? ) across it at some point.

[INewton]

Comatose,Thanks for pointing that out to me. That's what i was missing!. At this point i switched back to electrolytic capacitors and it worked fine. But i wanted to regulate the input down to 5 v. So i added a Seiko S-812C50AY-B-G (http://www.mouser.com/index.cfm?&handler=data.listcategory&D=*S812C50AYBG*&terms=S-812C50AY-B-G&Ntt=*S812C50AYBG*&Dk=1&Ns=SField&N=0&crc=true)
5 regulator from mouser.
The problem is at power up the voltage doubler can't double the input voltage. But if i disconnect the voltage regulator from the Vdoubler output and then connect it back on after power up the circuit works. How did you come up with a solution for this?
Thanks

Try a 1A diode from Vin to Vout of the doubler. It might raise Vout enough for the doubler to start working.

[rocky79]

Try a 1A diode from Vin to Vout of the doubler. It might raise Vout enough for the doubler to start working.

Thanks Newton,your a geniuous It did work! i used the 1N4148.
I am trying to understand why it worked? Thinking out load. Looking at the first page of the datasheet, it shows that there is an internal zener diode in place from Vout to Vin. So having an external one will double the amount of current that charges the output cap? that way it can drive bigger load?

[INewton]

Thanks Newton,your a geniuous It did work! i used the 1N4148.
I am trying to understand why it worked? Thinking out load. Looking at the first page of the datasheet, it shows that there is an internal zener diode in place from Vout to Vin. So having an external one will double the amount of current that charges the output cap? that way it can drive bigger load?

I was happy to help, but do me a favor, don't think out loud. :rolleyes:

[rocky79]

I was happy to help, but do me a favor, don't think out loud. :rolleyes:
:p But am i right?

[INewton]

:p But am i right?

Nope...

The diode pulls the output off the ground rail, which allows the internal switching (charge pump) transistors to get some voltage between emitter and collector.

[rocky79]

We use 10uf and even 22uf ceramic surface mount caps. They're not cheap, but they exist.

In fact, those ones ARE 10uf. But they're also y5v, which makes them wildly unsuited for this task. Y5V loses almost all of its capacitance(well over 90%) when you put a dc bias onto it. You'll need x5r or x7r grade caps to make the circuit work properly.

Digikey part number PCC2417CT-ND will work much, much, much better and isn't too unreasonably expensive.
for the V-doubler the datasheet shows the electrolytic cap with the polarity do you think a 10uf smd ceramic cap will work?
I am looking at the 581-tajb106k016 (http://www.mouser.com/index.cfm?&handler=data.listcategory&D=*581tajb106k016*&terms=581-tajb106k016&Ntt=*581tajb106k016*&Dk=1&Ns=SField&N=0&crc=true) from mouser

[Comatose]

That isn't a ceramic cap, thats a tantalum cap. They work nicely, just don't overvolt them or hit them with reverse polarity or they explode.

[rocky79]

That isn't a ceramic cap, thats a tantalum cap. They work nicely, just don't overvolt them or hit them with reverse polarity or they explode.
Is the positive terminal marked by the side marking? the cap datasheet doesn't even mention it

[rocky79]

Try a 1A diode from Vin to Vout of the doubler. It might raise Vout enough for the doubler to start working.
Newton,
with the diode in place the V-doubler lasted few days and now it burned. It was drawing 35mA. I was using the 1N4148. Anyidea what to do now?
Thanks

[INewton]

Newton,
with the diode in place the V-doubler lasted few days and now it burned. It was drawing 35mA. I was using the 1N4148. Anyidea what to do now?
Thanks

The 1N4148 is not a 1 Amp diode as I recommended. The startup surge killed it.

[rocky79]

The 1N4148 is not a 1 Amp diode as I recommended. The startup surge killed it.
the Diode is fine, it's the V-doubler that got burned.I will order a 1A diode for the new part

[Afroman]

I've never seen someone have so much trouble with a voltage doubler before. Looking at how much time you have spent on it I would just declare the chip as being crappy, and move onto something else. Maxim have many options, and they work great. Their free sample policy is good too.
http://para.maxim-ic.com/cache/en/results/5079.html
http://para.maxim-ic.com/cache/en/results/5080.html