Hi Stef. I see you are using tantalum caps. I have seen many posts complaining that these are subject to spectacular failure, taking out associated components (you can e ven see some on the Pass board by Nelson.)
Of course they are great because of low ESR and small size, and yes they are recommended on the data sheets. But I have changed to using low ESR electrolytic rather than tanatalum in view of dire warnings from others.
Anyway, your supply is simple enough to be simulated, I see you may have posted some sims, nice job.
Of course they are great because of low ESR and small size, and yes they are recommended on the data sheets. But I have changed to using low ESR electrolytic rather than tanatalum in view of dire warnings from others.
Anyway, your supply is simple enough to be simulated, I see you may have posted some sims, nice job.
The Tantalum reliability is well know but may be a bit exaggerated.
Caps from 70's was not reliable (and its depends). Today Tantalum are really reliable, specially polymer-tantalum commonly used in SMD design.
In fact, it's depend of the maker and price as many things. I've see Tantalum in product aged of 15 years, and more, still working fine. I've also see fresh no-name inexpensive Tantalum to explode.
Good supplier: AVX, Kermet
.
Caps from 70's was not reliable (and its depends). Today Tantalum are really reliable, specially polymer-tantalum commonly used in SMD design.
In fact, it's depend of the maker and price as many things. I've see Tantalum in product aged of 15 years, and more, still working fine. I've also see fresh no-name inexpensive Tantalum to explode.
Good supplier: AVX, Kermet
.
KEMET not Kermet. 
The LT3080 is now available as Sample for the ones who want to try it, but only as SMD MS8E. Will wait for the TO220 package.
The LT3080 is now available as Sample for the ones who want to try it, but only as SMD MS8E. Will wait for the TO220 package.
LT3080 resistor?
Hello! I just found the LT3080 and am considering for several applications and was looking at your schematic version 0.3 (as well as the LT spec sheet). The spec indicates that the reference voltage is set by forcing a 10uA constant current through Rset, (so for 15V, you want to Rset = 15V/ 10uA = 1.5M - perhaps a 1.2M resistor + 500K linear pot?)
Can you explain the reason for connecting R2 from output back to VR1 in the voltage set path?
Won't this have the effect of both shifting the voltage and possibly coupling any output noise back onto the reference point?
The way I read the datasheet, the voltage on the SET pin is the same as the output voltage. Thanks for any explination! - Ted
Hello! I just found the LT3080 and am considering for several applications and was looking at your schematic version 0.3 (as well as the LT spec sheet). The spec indicates that the reference voltage is set by forcing a 10uA constant current through Rset, (so for 15V, you want to Rset = 15V/ 10uA = 1.5M - perhaps a 1.2M resistor + 500K linear pot?)
Can you explain the reason for connecting R2 from output back to VR1 in the voltage set path?
Won't this have the effect of both shifting the voltage and possibly coupling any output noise back onto the reference point?
The way I read the datasheet, the voltage on the SET pin is the same as the output voltage. Thanks for any explination! - Ted
Re: LT3080 resistor?
Yes, you're true. The schem in 0.3 is not the best. It was an idea to use a small pot. No need.
1.5M + cermet 10 turns is better for an adjustable version. Or more to use 12v to 15v.
It's so bad that only the MS8E is available yet.
.
tlenthe said:
Yes, you're true. The schem in 0.3 is not the best. It was an idea to use a small pot. No need.
1.5M + cermet 10 turns is better for an adjustable version. Or more to use 12v to 15v.
It's so bad that only the MS8E is available yet.
.
Yes, I agree that the surface mount packages are quite difficult for the DIY community. Unfortunately, that seems to be what the high volume users of these parts want 
From the web site, it doesn't look like there is a thru-hole version on the horizon...
I'm actually more of a thinker than a do-er, but have been thinking (for a long time!) about building a regulator for the pre-driver section of my old Adcom 555-II. Unfortunately, this needs to have voltage in the range of +/- 50-60 volts, so this is one of the first single package regulators with low noise that will handle the high voltages required..
I'm unclear where the limitation for 60V on the output applies if there are not really any terminals connected to ground, but that's not an issue for non-power amplifier applications.
Looks like you may need to jump into the world of surface mount devices if you want to try this
Keep us posted on what you end of with!
- Ted
From the web site, it doesn't look like there is a thru-hole version on the horizon...I'm actually more of a thinker than a do-er, but have been thinking (for a long time!) about building a regulator for the pre-driver section of my old Adcom 555-II. Unfortunately, this needs to have voltage in the range of +/- 50-60 volts, so this is one of the first single package regulators with low noise that will handle the high voltages required..
I'm unclear where the limitation for 60V on the output applies if there are not really any terminals connected to ground, but that's not an issue for non-power amplifier applications.
Looks like you may need to jump into the world of surface mount devices if you want to try this
Keep us posted on what you end of with!
- Ted
** correction **
oops! I was looking at the LT3010 page and not the LT3080 page 🙄 ! That one IS, indeed, shown with a TO-220 package, but not yet listed on the sample request. Suppose we all need to be patient!
- Ted
oops! I was looking at the LT3010 page and not the LT3080 page 🙄 ! That one IS, indeed, shown with a TO-220 package, but not yet listed on the sample request. Suppose we all need to be patient!
- Ted
Yes, we just need to wait. How long, I don't know. I've sent a mail to Linear but no answer.
No updated schematic.
No updated schematic.
As long as you are going to the trouble of using expensive filter and bypass caps, perhaps you would "snub" the diodes --
the LT1030 looks like a great, low noise regulator...good post regulator for a switching supply, and you can regulate down to "0" volts - with other regulators you have to bias the adjust pin down to do this.
the LT1030 looks like a great, low noise regulator...good post regulator for a switching supply, and you can regulate down to "0" volts - with other regulators you have to bias the adjust pin down to do this.
jackinnj said:
Sorry not understood.
For the C9/C20, what will be the best? An 1uF MKS2 Wima or an X7R SMD C1206 Kemet?
.
take a look at this article by Jim Hagerman --
http://www.hagtech.com/pdf/snubber.pdf
for the RSET bypass Linear has stated "up to 1uF" -- the LT1030 differs from 3-pin regulators in the way that the error amp is set up. the device is also susceptible to "board leakage" -- so don't allow it to get dirty.
I look forward to getting this device and testing it out when it is available in TO-220.
http://www.hagtech.com/pdf/snubber.pdf
for the RSET bypass Linear has stated "up to 1uF" -- the LT1030 differs from 3-pin regulators in the way that the error amp is set up. the device is also susceptible to "board leakage" -- so don't allow it to get dirty.
I look forward to getting this device and testing it out when it is available in TO-220.
I was just reviewing your schematic, having ordered a 40 pieces LT3080 today. One question: aren't your LEDs installed backwards?
jwb said:
Yes, this is corrected.
I finally realized a prototype of an +-15 v psu using the LT3080 (see the schematic). It's working very fine (stay on for 72h). With 150mA load, the regulator are 32°C. Noise is very low (below my scope possibility).
http://www.awdiy.com/uploads/pdf/LT3080-PSU-DIY-1.0.pdf
The psu needed to be loaded or the LT3080 will not start (the LEDs are ok). For best regulation, the set pins and associated component need to be wired according to the datasheet (as a kevin sense connection).
I've see one or 2 times a problem at start with voltage reg (start à 14.5v and goes in 2 minutes at 15v, and stay). I don't know if this a problem with my scope or something else.
Great regulator and easy to use. I hope that Linear will go with a dedicated negative version (will sign the dead of the LM317/337 couple). 😉
Stef...
ps : I never received the Samples and had to order 25 at Digikey.
This will work -- and you don't need two transformers -- you can use your Linear LT3080 instead of an LM317 if you prefer.
An externally hosted image should be here but it was not working when we last tested it.
jackinnj said:
Interesting. The Bandwidth is 48Khz only?
I tried a look on National for the OPA5xx series. The OPA541 do not have a nice Package. Do you have another model to advice?
FOR MODERATOR: can we have this Thread migrated to "Power Supply"?
I haven't checked the BW vis a vis the LT3080 or the LM317 -- the open loop gain of the OPA541 is over 1 MHz.
The opamp shown is a Texas Instruments part -- formerly Burr Brown -- there are several in this family -- high power opamps.
The opamp shown is a Texas Instruments part -- formerly Burr Brown -- there are several in this family -- high power opamps.
My first miniVreg using the LT3080. Adjustable +0-20v with only one resistor.
Same size as a TO220. 78xx serie pin-out. 250mA without heatsink. 1A with heatsink (the back side is flat without component).
.
Same size as a TO220. 78xx serie pin-out. 250mA without heatsink. 1A with heatsink (the back side is flat without component).
An externally hosted image should be here but it was not working when we last tested it.
.
- Status
- Not open for further replies.