Author Topic: LT3080 High Current Equivalent for 4 channel bench supply  (Read 2838 times)

0 Members and 1 Guest are viewing this topic.

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
LT3080 High Current Equivalent for 4 channel bench supply
« on: September 24, 2018, 01:34:08 am »
Hello everyone,

Sorry if this topic has been previously addressed, but I'm in the process of building a 4 output Linear power supply and I want to use Dave's uSupply design as a reference.

In my design I have 2 low voltage, high current outputs (0 - 20v @0-5A each -- Call this channel A) and 2 high voltage, low(ish) current outputs (0-48v @0-2.5A each -- Call this channel B). I also want to make sure that I can put both outputs in either channel A or B in series or parallel to get more power if I ever need it. I already have 2 transformers for the supply, one 24v 11.6A one and a 48v 5.3A one. Now I just need to find a proper deign that can handle this power.

I was originally going to use this design for each of the 4 channels with an added capacitance multiplier filter to reduce noise: https://bit.ly/2pwLdjH

But the LM723's are wayy obsolete  |O and the only place I can find them is on ebay. So I was wondering if any of you guys know of a better design that I can use that will still fit my needs and that will give me the option to add a MCU for more accurate control later on if I ever feel like it.

Thanks  ;D
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9234
  • Country: us
  • $
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #1 on: September 24, 2018, 01:39:16 am »
I don't know if I would add a capacitance multiplier as a noise filter to a lab supply. Maybe to some op-amp rails in some kind of sensor or amplifier box.

But a good solution would be to use the LT3083 and its datasheet application notes, you can parallel them or mod them for high voltage operation.

Just test it first for stability on bad loads. It looks like you will need 2-3 LT3083's per 20V rail and 1 per each high voltage rail.

30$ of regulators for a 4 channel lab supply is not really that bad. Simple might be better because you just have 1 IC to replace if something fires, maybe 2. If you keep a few extra on hand and put them in the enclosure with some antistatic you can ensure uptime. Nothing worse then broken lab equipment.

or this obsolete thing
http://www.analog.com/media/en/technical-documentation/data-sheets/1038fa.pdf
« Last Edit: September 24, 2018, 01:45:44 am by coppercone2 »
 
The following users thanked this post: FotatoPotato

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #2 on: September 24, 2018, 01:49:43 am »
I did take a look at the LT3083 but the TO-220 versions are $10 a pop and If Im going to be using 8 - 10 of them then the cost will add up quickly. Is there any way to just use one and then add an external BJT to increase the power output capabilities and reduce the amount of them that I need to buy. I'm thinking something like this:

Also, How would I go about modding them for the high voltage operation. I'm assuming it would be the same as what I showed above but I'm not quite sure.

Anyway, thanks for the quick reply.  :)
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9234
  • Country: us
  • $
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #3 on: September 24, 2018, 02:08:15 am »
well you can also mod the lt3080 for high current, it has a diagram in the datasheet.

I think a transistor buffer is gonna be a bit quirky compared to 2 identical regulators in parallel but I won't do an analysis, you can probobly dig up dirt on both circuits though.

maybe use 2x of the lt3083's for the HV rails modded, then 2x lt3080's with transistor for the lower voltage rails

but scroll down in the datasheets for lt3080 and lt3083 and look at the schematics


also read the datasheet. THey have something like 80% match with good parts, meaning if you parallel them with a extra margin (cuz its a lab supply), you get like a 60% boost, so 3+0.6*3=4.8A

that means 6 lt3083s will do it, but thats 60$.

if you use the transistor load sharing circuit with the lt3080 you can get by using 2x lt3080 and 2xlt3083

reading datasheet carefully might tell you if lt3083 can transistor shared too. you can probobly also parallel them with the HV mod

Also if you parallel them I think you reduce noise by the square root of devices you add (has diminishing returns after like 4 that are pretty severe). DOn't know how to analyze the transistor one off the top of my head at this hour.
« Last Edit: September 24, 2018, 02:19:47 am by coppercone2 »
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #4 on: September 24, 2018, 02:27:24 am »
Looking at the Datasheet for the LT3080 I saw this diagram (Linked Below). I will probably replace the MJ4502 with a TIP142 due to cost and the fact that it has a much higher gain that the original part. I'll definitely do this for the 2 low voltage outputs. I also saw that the LT3083 has a high voltage mod (again, linked below). So it seems as though my problems have been solved by just reading... :palm: (I should do that a bit more lol)

I do have some concerns about the High voltage part though because 48v AC doesn't translate to 48V DC. Its more around 68V DC with no load and about 60-65v at max load. so that will be a problem but I'll just have to do a bit more digging.

Thanks for the help, I really appreciate it. Hopefully things work out well and I can get to using my supply.

Thanks  :-+
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #5 on: September 24, 2018, 05:10:40 pm »
I just took another look at the LT3083 and I saw that its output is also limited to 23V :palm: Now I do know that you can increase the input voltage using a mosfet on the input as I showed in the pic's posted above but how would I go about increasing the output voltage? I'll keep looking around for other variants of the LT3080 that support a higher voltage but as of now I'm kind of stuck.

Do any of you guys have any advice?

Thanks :)
« Last Edit: September 24, 2018, 06:33:47 pm by FotatoPotato »
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9234
  • Country: us
  • $
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #6 on: September 24, 2018, 06:30:53 pm »
yea stop posting at 4am i dont remember what my line of thought was
at least i solved half your problem lol
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #7 on: September 24, 2018, 06:33:07 pm »
I posted that post at like 1pm... Not sure why it thinks I posted it at 3am... weird
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9234
  • Country: us
  • $
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #8 on: September 24, 2018, 06:38:49 pm »
no I mean I was up late sleepy still posting

have a look through these they have nice features
http://www.analog.com/en/parametricsearch/11459

called the LDO+ series, like the LT3081.

http://www.analog.com/en/parametricsearch/10996#/d=sel|5422|s5|5351|5349|5417|5418|5573|5574|5575

search by voltage then see.
« Last Edit: September 24, 2018, 06:42:52 pm by coppercone2 »
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9234
  • Country: us
  • $
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #9 on: September 24, 2018, 06:55:38 pm »
if you want to float the LT3080 for high voltage, there is a nice circuit on page 698 in "The Art of Electronics", 3rd Edition. Figure 9.113

I am not posting it because GIMP is a piece of shit.
« Last Edit: September 24, 2018, 06:57:19 pm by coppercone2 »
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9234
  • Country: us
  • $
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #10 on: September 24, 2018, 07:00:53 pm »
Are you sure the circuit in the data sheet does not work?

The zener diode should limit the voltage drop across it, the thing floats anyway, so the output does not care what is going on other then in the input. If you look at it, there is nothing connected to it other then capacitors.

 I am on vacation tho
 

Offline Bassman59

  • Super Contributor
  • ***
  • Posts: 2501
  • Country: us
  • Yes, I do this for a living
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #11 on: September 25, 2018, 10:10:27 pm »
I did take a look at the LT3083 but the TO-220 versions are $10 a pop and If Im going to be using 8 - 10 of them then the cost will add up quickly.

... and this is why a quality bench power supply is not inexpensive.
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #12 on: September 26, 2018, 02:52:56 am »
I guess so, if you want the quality then you gotta pony up the cash for it  :-//
 

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #13 on: September 26, 2018, 03:19:08 am »
I think I have set my expectations a bit to high because 48v is just gonna cause unnecessary hassle, I think for the high voltage outputs I’ll keep it at 30V and just pump up the current to like 6A per channel. Other than that I just need to figure out hot to change Daves design to allow the greater voltage output and current output adjustment and I should be good to go. Fingers crossed
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9234
  • Country: us
  • $
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #14 on: September 26, 2018, 11:35:19 am »
that circuit should work. I figured out how to use gimp so I can get you what they wrote. But buy the book.

you owe me for this because Gimp IS FUCKING IMPOSSIBLE TO USE JESUS CHIRST



BUT THAT one can do 1-500V
 
The following users thanked this post: FotatoPotato

Offline FotatoPotatoTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
  • It's probably in reverse...
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #15 on: September 26, 2018, 02:04:54 pm »
Thanks so much  ;D This helps a lot. I’m definitely gonna buy volume 3. I have the 2nd edition but I haven’t seen it in a while (I should probably go find it and start reading again). Just one other question. If im going to have my high current channel be 0-12v at 10A, will a single 25A PNP BJT added to the LT3080 be sufficient? Or will I need 2 BJT’s parallel.

Btw I’m referring to pg. 17 Immage #1 on the datasheet

Thanks  :-+
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9234
  • Country: us
  • $
Re: LT3080 High Current Equivalent for 4 channel bench supply
« Reply #16 on: September 26, 2018, 06:11:58 pm »
I'm not sure, I think you have to keep base current in mind.

Test this stuff on a bread board though. I never made this circuit and I don't know what will happen.

Also use LTSPICE to test it, the lt3080 is in there pretty sure, you can see what the voltages and currents are so you can get an idea of what parts you can replace. It's free and actually easy to use.

Also, keep in mind the LT3080 and its variants have a thermal over ride. If you use a external transistor you don't have anything.

I think you can attach the LT3080 on the same heat sink to act as a probe to shut it down, but it might not be fast enough, so doing this might just prevent it from making a burning hot heat sink.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf