Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

June 02, 2023, 11:28:48 AM

Login with username, password and session length

Recent Posts


testing linear voltage regulator(s) input current limit?

Started by saturated, April 11, 2023, 07:15:32 PM

Previous topic - Next topic


hello all
I have a realistic power supply 13.8v 3A
I want to hook up and test an old 7808CT removed from my amp.  I dont know much but I am thinking 3 amps is gonna blow this thing up or melt it....
(I dont want to reuse it this is just me trying to learn some stuff...)
I also have a topwon 6303A that I havent unpacked yet maybe I can dial current down..
also I read the data sheet on some of the semiconductors...about a I could find was the output current.
I did some preliminary diode type tests per youtube...it seems to be good idk....if its good its good I will replace it regardless that dag blasted glue had to come off anyway.  Anyhow mucho beer and pizza to all I really appreciate it all the help I have gotten here.


The 7808 has an internal current limit. If it draws 3 Amps, that current limit isn't working and the chip is bad. To get to the current limit, you need a load resistor on the output, 8V/R = 3A. R needs to be 1.666 Ohms. Looking at the data sheet, With a 13V input the current limit at 25C is about 2.25A. 8V/2.25a = 3.555555 Ohms. Note that the current limit goes down as temperature increases. At 125C the current limit at 13V input is only 1.7 Amps.

Edit: I should mention that you need to use a heat sink when testing with any load current higher than about 100mA. The data sheet says the chip is protected, but they will fail if abused.

Edit2: You can use a small heatsink with a plastic "Vice Grip" ( made for holding wood as glue dries ) that has rubber jaws.


The 3A rating of your power supply is the max. it is capable of delivering.  It will only put that much current out if the load demands it.


thanks guys it is helping to actually do stuff in real life instead of just calculating current and voltage drops. 

just an FYI of my thought process....last night I had wondered if I could connect it to a 12v battery (automotive) immediately I was horrified imagining 650-700 CCA vaporizing the little device and kinda breathed a sigh of relief that I didnt do that.

so then this morning on the way to work I got to thinking...in my boat I have a big ass dedicated 12v deep cycle hooked to one thing...a Lowrance sonar.  and its not drawing hundreds of amps or glowing red from current. (lots of times I run sonar with the outboard off and so I have a separate cranking battery)

so I kinda came to my senses about this...I think my thinking got corrupted by imagining a variable power supply could dial in any current...maybe it can idk..

so anyhow I am going to take a 12v batt or power supply and connect to pins 1 and 2....then connect multimeter to pins 2 and 3 and see if it reads 8v.

also I plan to connect the DUT to a heatsink before testing

it also occured to me that this is precisely what Mr Tassie Viking instructed me to do while it was in the amp (thank you sir!)  :)

hey guys when people at Peavey dumped hot glue on components was that supposed to act as a heat sink?

thanks again


If a component is faulty it might be shorted out, then it can absorb a lot of power and go up in a puff of magic smoke.
With a 12v car or boat battery connected to it it can pull enough amps to make it burn up and even catch fire, even the leads can melt if they are not big enough.

If you place a resistor between the battery and component the maximum amps it can draw is calculated by I= E/R, or amps = voltage divided by resistance. That is the most amps you are allowing to pass through the circuit in case the component is faulty.

If you used a 6 ohm resistor between the battery and component the maximum amps will be I=12(volt)/6(ohm) which is 2 amps.

Of course the resistor needs to be able to survive the amps you pass through it, in case it is a shorted component it will be 12v and 2 amps. Power=volts x amps = 24 watts, I doubt you have a 24watt or bigger resistor laying around.

Another easy way to limit the amps is to use a light bulb instead of a resistor, in this case you would use a 12v globe from a car maybe.
I would use a small wattage light bulb to start with, something like a 4 watt tail light bulb would be fine, if the component is good there will be hardly any current used so the bulb will not light up and you would get 8 volts between pin 2 and 3 on the 7808 regulator.
If the regulator is faulty then the ligh bulb lights up if the 7808 is shorted out, or if it has gone open circuit then nothing will happen as there will be no way for the power to go through the regulator.

If you make a triangle over the formulas and then cover the one you want to work out it shows how to do it, I better explain better.

   V         V         
I          I
   P ,       R ,   

I=V/P,  V=I x P,  P=I x V,      I=V/R,  V= I x R,  R=I x V

The light bulb is there to protect not so much the 7808 you are testing but the battery and the wires you are using to test the 7808 with, and it also stops you from breathing in too much smoke.

The hot glue was used to keep the components steady and safe from vibrating too much, if you wiggle and bend the component leads too much they will break.

Large capacitors used to be glued down a lot because with their weight, and if someone playing a 100 watt combo amp at full blast there will be some vibrations if you are doing it right. Also Amp heads often get placed on top of the speaker cab as well.
Imagine a 400 watt Bass amp speaker at full blast, sit on top and you will sing in a natural vibrato.

There are no stupid questions.
There are only stupid mistakes.


well hell.....
I hooked everything up...wish I could post some pics

as soon as I turned the power supply on my meter was showing 8.03V
so I guess this little guy can work when its not soldered in the amp...idk smh  >:(
so back to the drawing board...or schematic...
on the plus side this was lots of fun damn I learned a lot in the last few days....
just wish I had a different cadaver to pract8ce on because I really love this amp  :)

so thanks guys dont bail on me yet ha ha


hello ssguitar friends

since this device seems to test good out of circuit I am venturing that I did not measure it correctly in circuit....I dont remember where I had ground..I likely clipped it to the metal chassis...
anyhow my new rabbit hole is going to be learning about ground(s)

thanks again


It can put out the proper 8V unloaded in your test circuit, yet the voltage could collapse with a load on it.
In your test circuit, put a resistor from the output to ground.  Something in the 100 to 470 ohm range should do the trick.


thank you sir that thought certainly did occur to me much like many things that test ok without a load and fail with one.
I am ordering a couple new 7808 but dragging my feet trying to assemble a large Mouser order to offset shipping.
I do not have any resistors on hand but def have some junk laying around where I can desolder one and use it.


alright guys I desoldered a resistor a d measured it at 1 kohm....seemed maybe too high.
I then made up some jumpers from odds and ends...hooked everything up..put my multimeter on each side of the resistor...8.05v....dang it  >:(

so i am extremely interested in seeing if I can get a good ground on that center pin when the replacement is soldered back in circuit.

frustrating..but this has been a good experience...thanks for your help
You cannot view this attachment. 


Quote from: saturated on April 15, 2023, 12:00:41 PM...hooked everything up..put my multimeter on each side of the resistor...8.05v....dang it  >:(
One end of the resistor needs to go to ground and would measure 0 volts.


this is how i had it...exception being I had grounds hooked up to the tab on top...idk  :P

You cannot view this attachment.

thanks for your help


Ok.  Set-up looks correct.  I guess you meant you had one meter probe at each end of resistor.
I thought you had meant you checked from each end to ground.  My mistake.

A 1K load on the 8V line only tests at 8mA which is still pretty low current.  It's probably alright, but could still be bad.
100R resistor would test at 80mA which would be a better test, as the part should be able to deliver 1A of current, though that much would require heatsinking.


yes sir thanks for your help
I appreciate you checking up on that

I need to get those new ones in thats for sure sorry for the delay

I actually found a box of old NIB RS components in a thrift store today and sifting through it was an LM317  :)  :'(


I reckon you guys knew this all along but I think with your help I learned a lesson here....

devices can appear to test "good" out of circuit and still be bad