Why is the fuse blew out? (2)

[psychepool]

https://ampgarage.com/forum/viewtopic.p ... 23#p400123

Lately, I posted something like this.
It was a question about the problem of a fuse blew out when a power plug is connected even when the power is turned off.

The last thing I tried on that post was a replacement of rectifier diodes. Of course got a same result and nothing has been resolved.


I've been so busy for a while that I couldn't tried anything else.
Many peoples said that I should make a light bulb limiter, but I didn't have time to do that, so I checked the status of the power transformer in a different way.

With all center tabs of my transformer have been grounded in the chassis, I disconnected all other power supply lines from the circuit.
Then I put on a new fuse and plugged in the power plug.
Wow! The fuse didn't blow out!

I measured the voltages of several lines on the secondary side.
About 330 VAC was checked on the B+ tab, 3.2 VAC on the heater tab (specifications with center tap), and 6.3 VAC on the power tap for relay.
The Transformer itself confirmed normal operation with no major errors, according to the specifications I have ordered.

Then, I tried connecting to the circuit one by one from the tap with low voltage.
First, I connected the heater tab to the tube socket. The fuse is working normally and the voltage is good.
The power tap for the relay was then connected to the board. Works perfectly.
Finally, the tab for B+ was connected to the rectification diode. This is where the fuse blows!

No problems on the other tabs.
Nothing was connected the otherside of the diode.
So, now I can think of a problem only between the power transformer and the rectifier diode .


But... why is it?
I think it's very unlikely that only the diode I used happened to have been defective.
However, since there is no reason to think about any problems other than diodes, I will try to connect the new other diode again.

I I tried it differently this time because reconnecting to the board's diode will result in the same result.
I connected the UF4007s directly to the B+tap.

06.jpg

I tried connecting the power plug carefully, paying attention to the electric shock. Oh! The fuse doesn't blow!
Why is it? It is a difficult situation to understand.
I looked for a moment and checked the voltage.
And, another problem occurred.
After rectification, the voltage was only 300 VDC.
Wow... the problem happens endlessly.


The writing has lengthened.
to sum up
1. The fuse-blowing is only occurs when the B+power tap is connected to the rectifier diode of the board.
2. But when I connect the rectifier diode to the B+Power tab directly, the fuse did not blow out.
3. However, after rectified, the voltage is less than 100V below the normal value expected.

Where can I find the problem?
Please help me.


Attached is a photograph of the current state for reference.
The soldering is not very neat because of many re-soldering.

02.jpg

03.jpg

04.jpg

05.jpg

The chassis ground is grounded in one location: multi-electrolytic cap for preamp, bias circuit, powertrasformer secondary main center tap, output transformer secondary 0 ohms tab, output jack ground.
Power socket, heater center tabs, main filter caps (100uF+100uF), and the ground of the relay power supply are connected in separate locations.

[martin manning]

I think you have proven that the diodes on the board are shorted. The DC voltage reading you are getting with the new diodes connected is low because you don't have a reservoir capacitor connected. See the waveforms shown on this page: http://www.valvewizard.co.uk/bridge.html

[psychepool]

I think you have proven that the diodes on the board are shorted. The DC voltage reading you are getting with the new diodes connected is low because you don't have a reservoir capacitor connected. See the waveforms shown on this page: http://www.valvewizard.co.uk/bridge.html

Thank you for your reply.
By the way, most amp that I ever made has the stanby switch at the front of first filter cap(reservoir cap).
Therefore, in the standby mode, the rectifier was not connected to anywhere except the power transformer, but always the voltage was normal.

As I have seen the fuse blowing continuously, I'm worried that if I connect high voltage to the capacitor while it is not yet normal, the capacitor will broke.
Will connecting the rectifier to the cap solve the voltage problem?

If this solves the voltage problem, how do I connect the rectifier diode?
As you can see from the pictures above, There have not found a short or other problematic part.
For now, the only attempt seems to be to remove the existing turret pin and put a new turret in the new position. If you connect the rectification diode to the cap and get a normal voltage...

Anyway, thank you for your answer!

P.S.
Now I found and read this article.
https://ampgarage.com/forum/viewtopic.php?f=6&t=32210

I think the thread mentioned in the beginning of this article is just like the thread I wrote.
I think it's fortunate that it was inferred in roughly the same way, although not exactly that way.

[psychepool]

I tried connecting the end of the rectifier diode to the reservoir capacitor.
I got 443VDC!

Most of the amplifiers I've made so far were not connected rectifier end to reservoir cap, and I remembering there's normal voltage.
But may be it was just my false memory.

Now I need to reattach the rectifier diode to the board,
But I can't find a problem with what's on it, so...I'll remove the currently installed turret pin, and install the turret pin in the new position and wire the diode.

I'll leave a result here after the do it.
Thank you for your help.

[martin manning]

I tried connecting the end of the rectifier diode to the reservoir capacitor.
I got 443VDC!

Of course!

Now I need to reattach the rectifier diode to the board, But I can't find a problem with what's on it, so...

Diodes can have a soft failure, where they will test ok at low voltage, but fail to prevent current flow at high reverse voltage. I suspect that is what you have.

[psychepool]

I've solved it.

In fact, could I call it a solution? I don't know why it's settled.
And I've done the same thing before, and I don't even know why I was in trouble.

Anyway,
As shown in the picture above, it's working without any problems with the diode is directly connected to the 325 VAC tap.
So at least in theory, there should be no problem when this is fitted to the turret board as it is.

Before attempting this, I was reconnected the 325 VAC tab to the diode that had previously been mounted on the board and the diode connect to the reservoir capacitor.
The fuse blew again. I hoped it would be solved easily, but I had to do a troublesome task.

I removed the diodes on the turret and removed the turret pin as well.
Then I install turret pins on the new points and connect the previous diodes that normally works.

00.jpg

Connect the 325 VAC tab, plug it in, and power it on.
Wow! The fuse won't blow and it's working!
But I found that the ground of the capacitors wasn't connected to the chassis. The DC voltage was only about 300V.

I connected the ground of the capacitor to the chassis and tried again.
But the voltage was measured at 0V.
When I checked the fuse, it was blown.
Why is it?

I didn't have any more ideas to try, so I just put on a new fuse and turned it back on.
Normal voltage was measured!!(around 440 VDC)

Thinking that it might be operating temporarily, I repeated measuring the voltage 3~4 times after unplugging it, putting it back in, and raising the power.
Fortunately, there is no problem anymore.



Why did it have this problem?
I think it's just a coincidence that the diodes are bad.
I guess it got too much heat from tight assembly, but that seems to be a low probability.

It is also mysterious thing that another fuse exploded in the last successful attempt.



First of all, it's working well even with a few retries, so I'll move on to the next level.
I just hope nothing goes wrong until it completely done.

[pompeiisneaks]

So if I'm understanding the build you have there, you should NOT ground the rectifier. It's a Full wave (FW) rectifier as connected. This doesn't get any ground. If you needed higher voltages, then you may have wanted a Full Wave Bridged rectifier (FWB) which requires you ground where you've input the high voltage, and then connect the high voltage to the center wires. Do you have a schematic of the amp as it's supposed to be? Basically you have it wired right now correctly, do not earth it, or you're sending the full B+ to earth and will cause a short that will either blow a fuse or your transformer.

This article explains the two types and has schematics that show the two.

https://www.electronics-tutorials.ws/diode/diode_6.html

I hope that makes sense? The full wave bridged will increase the voltage 1.4 times what you input, and I think full wave does .9 (probably wrong, going from memory) but if your voltages were 300 with the FW, then with FWB you'd get about 460VDC out of the FWB, where you'd connect the two red wires in between the middle of the two diodes and ground the two entry points you're using right now. I'd be careful connecting it to anything on the output of the FWB until you've tested what voltages it's putting out unloaded to ensure you don't blow up any cap's you have after it.

Edit: I just noted your first filter cap is a 350VDC cap, which sounds to me like the amp want's under 350V for that first node, so you're probably fine
~Phil

[sluckey]

Edit: I just noted your first filter cap is a 350VDC cap, which sounds to me like the amp want's under 350V for that first node, so you're probably fine

Look closely and you'll see that the first node is a 100uf/350v in series with another 100uf/350v cap, for a total of 50uf@700v.

[pompeiisneaks]

Edit: I just noted your first filter cap is a 350VDC cap, which sounds to me like the amp want's under 350V for that first node, so you're probably fine

Look closely and you'll see that the first node is a 100uf/350v in series with another 100uf/350v cap, for a total of 50uf@700v.

Ahh you're right, so yeah, if that amp wanted more like 460VDC, then you've definitely got it wired wrong. (that black wire at the bottom was hiding when I first looked)

~Phil

[psychepool]

Thank you for your reply.

I know that fullwave bridge rectification exists.
But I've never tried it when I making an amplifier before.

This is the full circuit diagram of the amplifier I'll make.
There may be places where minor errors have not been corrected because it's intended to be seen personally.

01.jpg

This is the configuration of the previously planned power supply / rectifier part.

02.jpg

I always make a diode rectifier like this.(serial connection from high voltage tap)
And in most cases, and the standby switch was located before the smooth capacitor was connected.

As I recall, when I wired like this, the expected voltage(approximately 440 VDC) came out even in a standby state(not connected to a smooth capacitor).
Maybe my memory is wrong because I don't make amps often.

Now it changed like this.

03.jpg

Connect the rectifier diode to the first smoothing capacitor (100uF+100uF series) and then connect the standby switch.
This results in an expected voltage of 440VDC.


The question is, no matter what the above-mentioned method of connection, no fuse has ever exploded.
More mysterious thing is that the new tried and working condition is also theoretically the same as the previous fuse blowing state.
Was there an unwanted short on the board? But when tested in meters, there's no short anywhere.



P.S.

There need a correction in the circuit diagram. The ground of 100/100uF is seperated with other caps, and connected to the chassis

[pompeiisneaks]

So yes, your amp needs just a Full Wave rectifier. Do not ground the rectification area, you DO need to ground the center tap of the transformer though.

Is that how you had it connected when it blew the fuse?

~Phil

[psychepool]

No, I didn't connect any parts of the rectifier diodes to ground or chassis. Just connect the center tap of the power transformer to the chassis.
It is as shown in the circuit diagram.
The condition when the fuse blow is the same as the second image of the above post.

[pompeiisneaks]

Sorry I misread this:

I connected the ground of the capacitor to the chassis and tried again.
But the voltage was measured at 0V.
When I checked the fuse, it was blown.
Why is it?

as you connecting the rectifier to the capacitor ground. I would basically set your DMM to continuity mode and test anywhere that's part of the B+ to ensure nothing is showing route to ground. I've found a couple different builds where I did this accidentally via this method. one time I had a solder blob due to my soldering still sucking that had reached down from the board to the ground. Another time I had an under board wire I accidentally pinched between the standoff and the board, and had to replace it. It's a great way to find the short. To me it definitely sounds like either a short or at least a very low resistance path to ground where there shouldn't be one. I don't know that the diodes are bad, but to test them set the DMM to diode mode, one direction you should see a voltage drop and the other should do nothing. If they test fine, then they're not the problem.

~Phil