LM 1084/338/317/etc... voltage drop at < 1A

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
solar3000
Posts: 31
Joined: Fri May 10, 2013 1:55 pm
Contact:

Re: LM 1084/338/317/etc... voltage drop at < 1A

Post by solar3000 » Wed Jan 03, 2018 11:23 pm

OK so did the suggested measurements.

Just to humour myself, I measured the bare transformer at 60 Hz. I measured at the point after the rectifier and mister meter says 120 Hz.

The bare transformer measures voltages like this: CT is 16 volts. Entire secondary is 34 volts. So would you say this is a 24/25 volt transformer?

User avatar
haklesup
Posts: 2843
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: LM 1084/338/317/etc... voltage drop at < 1A

Post by haklesup » Thu Jan 04, 2018 10:25 am

If you measures 34V peak to peak, it would be 24VRMS (34*0.707)

120Hz is expected but says nothing about the current you are sourcing from the transformer. your probablem may not be voltage, it acts like you are running short of current. So if the transformer is 5A ish then maybe the diodes or filter circuit are restricting the current but it looks sufficient at first blush.

If it were me I would leave the center tap open and use a full wave bridge rectifier but it should be equivalent.

you have LM1084 Are they in the TO-3 can? Those are impossible to get now

solar3000
Posts: 31
Joined: Fri May 10, 2013 1:55 pm
Contact:

Re: LM 1084/338/317/etc... voltage drop at < 1A

Post by solar3000 » Thu Jan 04, 2018 7:49 pm

haklesup wrote:If you measures 34V peak to peak, it would be 24VRMS (34*0.707)

120Hz is expected but says nothing about the current you are sourcing from the transformer. your probablem may not be voltage, it acts like you are running short of current. So if the transformer is 5A ish then maybe the diodes or filter circuit are restricting the current but it looks sufficient at first blush.

If it were me I would leave the center tap open and use a full wave bridge rectifier but it should be equivalent.

you have LM1084 Are they in the TO-3 can? Those are impossible to get now

In reverse...

My LM1084, LM338, and LM317 are to-220 type.

I just tried the full secondary and a bridge. The DC is 35 volts. Then with the capacitors, the voltage went up to 45VDC. I connected the full circuit for one second and the chip shut down. I think the voltage is way too high. Somehow I think the transformer is higher than 24/25/26 volts. I dug up an old transformer I got from the now dead Radio Shack. It is marked 25 V 2A transformer. With it thru a bridge and two capacitors, the measured voltage is 30 volts DC. And it was 27 volts AC vs the unmarked transformer of 34 volts ac.


Additionally, I tried this...
I found a switching power supply that outputs 5-16 volts DC at a max of 32 Amps. So I took out the connection to the caps, diodes, and transformer. And I connected this switching supply directly to the circuit. I had it set to its max of 16 volts and 32 amps. Seems to be the same problem. But max regulated voltage is about 10 volts. So I tested low voltage devices and regulated the voltage down to 5 volts.

I'll keep trying. I didn't melt anything yet.

User avatar
Lenp
Posts: 1287
Joined: Thu Oct 26, 2006 8:11 pm
Location: Maryland
Contact:

Re: LM 1084/338/317/etc... voltage drop at < 1A

Post by Lenp » Thu Jan 04, 2018 9:11 pm

FYI
The common transformer does not regulate. The label voltage is at the labeled current. At a lower current it will be higher voltage. At higher current the voltage will go down until the VA rating of the transformer is exceeded then the smoke gets out. I would expect the secondary voltage to be high if there was no load. Also the DC voltage better be higher than the A.C. or the theory of rectification has been wrong forever. Most designers today don't even design power supplies into their systems. A switchmode supply, with excellent regulation and a compact footprint is often cheaper than a power transformer, plus it has a list of approvals, and a warranty. For a DIY with a junk box of parts, that's another thing.
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

solar3000
Posts: 31
Joined: Fri May 10, 2013 1:55 pm
Contact:

Re: LM 1084/338/317/etc... voltage drop at < 1A

Post by solar3000 » Thu Jan 04, 2018 11:00 pm

Uhhh..huh-huh...problem solved.

same circuit.

I connected a 20W halogen bi-pin lamp one end to the ground (anywhere) and then the other pin directly to the output, even at the heat sink tab. The bulb lit up and didn't drop a tenth of a volt.

The real problem is that shiitie solderless board that I promptly placed in the garbage.

User avatar
Lenp
Posts: 1287
Joined: Thu Oct 26, 2006 8:11 pm
Location: Maryland
Contact:

Re: LM 1084/338/317/etc... voltage drop at < 1A

Post by Lenp » Fri Jan 05, 2018 5:37 am

From my first post......
"Possibly the breadboard connections are not perfect, or too small, check for voltage drop across the breadboard connections."
:mad:
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

solar3000
Posts: 31
Joined: Fri May 10, 2013 1:55 pm
Contact:

Re: LM 1084/338/317/etc... voltage drop at < 1A

Post by solar3000 » Fri Jan 05, 2018 6:48 am

yah. No more heat. The chip is just somewhat warm.

So, now it does not drop voltage up to 1.5 amps at a setting of 12 volts. With 2 amps, the voltage drops to 9 volts. So, here I go again.

Next step is to make a copper board or at least try a pre-made one like those from the dead Radio Shack.

I'm sure I'll be fine.

I dug up one I made when I was a boy. It works fine under .5 A. It uses LM317. Inside, I remember using wrapping wire and solder. Of course, if you remember the wrapping wire from Radio Shack ( RIP ), you'll remember it was very tiny like 30 AWG or slightly better.

User avatar
haklesup
Posts: 2843
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: LM 1084/338/317/etc... voltage drop at < 1A

Post by haklesup » Fri Jan 05, 2018 1:39 pm

the max Vin-Vout on all of these parts is 40V so you may be too close to the max. Furthermore, the greater Vin-Vout is, the more power you will need to dissipate in the regulator. to get the full 40V drop at full current you would need an exceptional heatsink. (just take the voltage drop times the current and that's how much heat you need to remove.)

I suggested series resistive connection somewhere, those reusable breadboards have notoriously poor connections, if you use a big wire, you often cannot use a smaller wire next time.

Try connecting the full wave bridge rectifier to the CT and one side, (leave the other terminal no connect) that should cut the voltage in half but keep the current sufficient.

solar3000
Posts: 31
Joined: Fri May 10, 2013 1:55 pm
Contact:

Re: LM 1084/338/317/etc... voltage drop at < 1A

Post by solar3000 » Fri Jan 05, 2018 5:12 pm

These just came in a hour ago.

https://www.amazon.com/gp/product/B00K8 ... UTF8&psc=1

Works great. No more hot capacitors.

Post Reply

Who is online

Users browsing this forum: Google [Bot] and 8 guests