Finding a short

[dragonworks]

I have only tinkered with amps that come on line but have problems. I now have a Gemini II that has a short and just blows the fuses. I don't don't know how to trace a short.I checked the PT and the OT they check out good.  I went from the positive sides of the filter caps to ground and found no conductivity, is this checking for a short?
I went from the B+ rails to ground and found no conductivity, is this checking for a short?
I fired up the amp without any tubes in it and still blew the fuse.  I looked for a video but could not find one. Someone worked on this amp. They recapped it, installed a three prong plug and kept the standby switch. When I first bought it, it would come on line, but it wouldn't come out of standby. I put it aside. Now she is blowing fuses.
I went on line looking for a video to help me out but could not find one.ThanxJ.D.

[sluckey]

Disconnect the PT leads from the bridge rectified. Still blow?

[dragonworks]

I can try that. Will have to take a look at the schematic and the wiring to make sure I have the right connections. Next time I am in the basement I will give it that test.
But still, I would like to know how to check for a short, was I doing it correctly on the rail and the caps? This is an ongoing problem I have. Simple basic testing techniques. I know ohms law, I know how tubes work,  how the passive components work, how the different sections of the amp work with each other, etc. I can basically read a schematic now, relating it to the chassis is the hard part?  What I don't know is, like testing for a short, what components can be tested in circuit and which ones can't. Which ones have to have one side lifted etc. I think all the videos I have been watching take it for granted that you know these things. A gent on a video on youtube told me after the three pronged plug was put in the standby switch is no longer needed? Where do I find the  basic basic testing techniques?ThanxJ.D.

[acheld]

A gent on a video on youtube told me after the three pronged plug was put in the standby switch is no longer needed?

Without being impolite, this sort of statement is why it is really tough to trust anything you find on the internet.

Where do I find the  basic basic testing techniques?

I'm not aware of a website that does that specifically.

That said, Sluckey just provided your first step in the process.  He is trying to isolate the problem, starting with the power transformer. 

[sluckey]

I can try that. Will have to take a look at the schematic and the wiring to make sure I have the right connections.

You don't need the schematic. Just look at the bridge rectifier inside the amp. There will be two wires from the PT that connect to the "AC" leads of the bridge. I can see it in the pic you posted. Unsolder these two leads and see if the fuse still blows. This is called divide and conquer troubleshooting. If the fuse still blows, look for the problem back toward the PT. If the fuse holds, look for the problem toward the bridge or filter caps. And since you have already checked the filter caps, the problem would most likely be the FWB.

Troubleshooting a fuse blowing problem can eat through a lot of fuses in a hurry. I suggest you build a light bulb limiter and learn how to use it. Saves a lot of fuses. See page 1 of this pdf...

     http://sluckeyamps.com/misc/Amp_Scrapbook.pdf

But still, I would like to know how to check for a short, was I doing it correctly on the rail and the caps?

Use your ohm meter to check resistance between point A and point B. If the meter reads zero ohms then the two points are shorted together. Don't use the beeper continuity function for this.

A gent on a video on youtube told me after the three pronged plug was put in the standby switch is no longer needed?

Now that's just plain stupid... or you misunderstood what he said.

[dragonworks]

This was his reply verbatim.
"With a 3-wire power cord, the polarity switch should have been removed from the circuit. You really need to seek assistance from a qualified tech."
It is a standby/polarity three way posistion switch which I had stated. He may be correct and missed the standby part. I think the polarity section of ithas been bypassed. The .05uf 1000v cap on the switch is no longer there. (see schematic, I have it circled in red along with others)

[dragonworks]

Before I do this: We are talking the two red wires coming out of the PT connecting to the diodes in the full wave bridge rectifier shown in the photo and schematic, correct?

[dragonworks]

How do like the size of the 20uf 600V electrolytic that was installed in there. I think I will find me a smaller one.I have a variax, can I use that instead of a light bulb limiter. I also have an isolation transformer that can be set from5 to 135 volts.

[sluckey]

Before I do this: We are talking the two red wires coming out of the PT connecting to the diodes in the full wave bridge rectifier shown in the photo and schematic, correct?

yes

[sluckey]

How do like the size of the 20uf 600V electrolytic that was installed in there.

I like it just fine. Own a pair myself.

I have a variax, can I use that instead of a light bulb limiter. I also have an isolation transformer that can be set from5 to 135 volts.

No. You need the LBL.

[dragonworks]

I will have to build a LBL. I disconnected the two red wires leading to the rectifiers and installed a 2.5amp fuse and she didn't blow. Now to the next step. As stated, I have checked continuity (0hms) from the positive sides of the filter capacitors, to the chassis, with my ohm meter using the ohm section, not the continuity section, and I can see no reading in ohms. But I am still not sure if that is checking to see if there is a short there. I was using the positive side of the filter caps as point A, and the chassis as point B. You stated, measure from point A to point B, which to a novice is  vague. I understood that to mean from point A, of the component or rail etc, that I am checking, to point B, being the chassis of the amp. These are the basics I am trying to pin down. Maybe they are not that basic.   The caps you installed aren't hiding a bunch of components and soldier joints under it lol.

[sluckey]

I disconnected the two red wires leading to the rectifiers and installed a 2.5amp fuse and she didn't blow.

That's good. It means the PT is probably OK, but let's make one more check just to be sure. Leave the PT red leads disconnected from the board an measure AC voltage across the two red leads. You should have about 350VAC. If so, the PT is good. What do you actually measure?

Now to the next step. As stated, I have checked continuity (0hms) from the positive sides of the filter capacitors, to the chassis, with my ohm meter using the ohm section, not the continuity section, and I can see no reading in ohms. But I am still not sure if that is checking to see if there is a short there. I was using the positive side of the filter caps as point A, and the chassis as point B. You stated, measure from point A to point B, which to a novice is  vague. I understood that to mean from point A, of the component or rail etc, that I am checking to point B, being the chassis of the amp.

You checked the filter caps properly. But don't get hung up that point B is always chassis because it ain't. Point B just happened to be chassis in this case simply because the negative side of the filter caps are connected to chassis.

But what if you want to check to see if a coupling cap is shorted? Neither side of the cap is connected to chassis. In this case, point A is one side of the cap and point B is the other side of the cap.

Bottom line... The term "short" or "shorted" does not always mean "shorted to chassis". Write that on the wall over your workbench.   

Now back to finding out what causes your fuse to blow. Now that we know your PT is not the culprit let's look at the next likely suspects... the diodes in the bridge.

First, how to check a diode... Set your DMM to diode function. Connect one probe to a diode lead and connect the other probe to the other lead. You should have a reading of about .5V. Now reverse the probes and you should read nothing. This is how a good diode reads. .5V one way, nothing the other way. This is the only good reading. Anything else means a bad diode.

Now to your amp...Unsolder one lead of all four diodes and lift the diode in the air. Now check all four diodes as stated above. If any of the diodes check bad, replace all four. Use 1N4007 or 1N5408 as replacements. Did you find a bad diode?

[PRR]

...
"With a 3-wire power cord, the polarity switch should have been removed from the circuit. ..."
It is a standby/polarity three way posistion switch which I had stated. .....

You did not say "polarity" until this quote.

Such switches are *usually* ON-off-ON, but the two ONs reference chassis to different sides of the line. NOT usually a standby function on a switch like that.

Yes, Light Bulb!!!

[dragonworks]

Before I go to the last step. I checked the output of the PT and it was 385 Volts. Then, with both transformer leads still disconnected, and all four diodes still soldered in, I checked them with the diode function. When looking at the photo I provided, the lower left one read .836, overload. The lower right one read 0,0. The upper right one read .832, overload. The upper left one read .791, overload. I checked the caps etc again with the meter set at the highest setting and I was seeing around .5 meg. I think I had the meter on a very low setting last time around and the PT was still hooked up. I didn't proceed to the next step because it appears the lower right one is bad. Should I replace all, and then carry out the next step?I want to thank all of you. This has been invaluable to me.I just dug four of my books out and I am going to read them again. Mostly on test procedures.
I just wanted to post this again for anyone who missed it and doesn't know about it. A great resource for free download of old electronic books, no strings attached. http://www.tubebooks.org/technical_books_online.htm  Thanx againJ.D.

{EDIT- fixed URL --PRR}

[sluckey]

all four diodes still soldered in, I checked them with the diode function.

YOU MUST DO WHAT I SAID!!!!!!!!!!!!!!!!!

The checks you did mean nothing.

[Willabe]

First, how to check a diode... Set your DMM to diode function. Connect one probe to a diode lead and connect the other probe to the other lead. You should have a reading of about .5V. Now reverse the probes and you should read nothing. This is how a good diode reads. .5V one way, nothing the other way. This is the only good reading. Anything else means a bad diode.

Now to your amp...Unsolder one lead of all four diodes and lift the diode in the air. Now check all four diodes as stated above. If any of the diodes check bad, replace all four. Use 1N4007 or 1N5408 as replacements. Did you find a bad diode?

[dragonworks]

Misunderstood, I thought you meant check them first while soldered in. I will go back down, unsolder them all and check them and relist the values.This is what I mean about the simple basics.Thanx againJ.D.

[dragonworks]

This time with diodes lifted, in same order, with the PT disconnected I get almost the same readings. Lower left, .823&overload, Lower right, 0&0, Upper right, .825&overload, Upper left, .791&overload. I take it from this that when checking diodes it cannot be done in circuit and one side at least has to be lifted? Is this always so?

[PRR]

I'm not following the thread but: four diodes in a ring, any one part you try to measure, there is always an other-way-round path to spoil your readings. (The 2 diode CT form is no better.)

Given an apparent short, not in the PT, I'd just replace the rectifier with prejudice. With a stronger part. Under-specced rects may work through hundreds or thousands of turn-ons, but eventually they go sppoooftt. They were cheap in 1979 and cheaper today, I've spent too much time fooling with blown rectifiers.

[sluckey]

Misunderstood, I thought you meant check them first while soldered in. I will go back down, unsolder them all and check them and relist the values.This is what I mean about the simple basics.Thanx againJ.D.

Simple basics... Read, understand, and follow instructions. If you can't do this, you will never learn this stuff.

[sluckey]

This time with diodes lifted, in same order, with the PT disconnected I get almost the same readings. Lower left, .823&overload, Lower right, 0&0, Upper right, .825&overload, Upper left, .791&overload.

There is a bad diode. Guess which one. Then replace all four.

I take it from this that when checking diodes it cannot be done in circuit and one side at least has to be lifted? Is this always so?

I already told you, it totally depends on the circuit. If you could accurately check the diodes in this bridge circuit, I would not have told you to lift one end of each diode.

[dragonworks]

the one reading 0 either way. The diodes are now on order and I will have to wait for them to arrive. Once I have them back in I  will return. Also ordered fuses. The diodes look original, there is no evidence of them ever having been replaced.I ordered ten 1N4007s.

[dragonworks]

In the meantime I will put together a LBL so when the diodes do arrive, it will be ready.Thanx all again."The journey starts with one step."

[HotBluePlates]

the one reading 0 either way. ...

Correct, that diode is shorted.

The shorted diode allows reverse-polarity voltage to get to the electrolytic filter caps, which causes them to become a short-circuit.  BAM, the fuse blows.

I know Sluckey can seem curmudgeonly, but there's a lot to be learned from how he steps you through a troubleshooting process.  Me, I've just about lost all patience with folks on another forum not following directions so I don't even try to help them troubleshoot any more.   

[dude]

Nice vocabulary HBP.   Yeah, Sluckey can be cantankerous but you’ll be pressed hard to find a guy with his patience to help one, probably why he doesn’t answer my questions anymore  ! He is the man here, puts in a lot of time helping solve rookies’s  issues, but if you should know the answers, watch out  .

[Keppy]

This was his reply verbatim.
"With a 3-wire power cord, the polarity switch should have been removed from the circuit. You really need to seek assistance from a qualified tech."
It is a standby/polarity three way posistion switch which I had stated. He may be correct and missed the standby part. I think the polarity section of ithas been bypassed. The .05uf 1000v cap on the switch is no longer there. (see schematic, I have it circled in red along with others)

The switch is a two pole switch. One pole switches the death cap from one side of the PT primary to the other. This is the polarity function. Without the cap, this half of the switch does nothing.


The other pole switches the connection from PT center tap to ground, controlling the actual standby function, and is unaffected by the addition of the three prong plug. The outer lugs are wired together, so both "on" positions are exactly the same.


Thus the switch is still useful as a standby, though you could get the same operation from a simpler single pole on/off switch. There's no reason not to use the existing switch though, except that people will needlessly wonder if the polarity matters.

[dragonworks]

"YOU MUST DO WHAT I SAID!!!!!!!!!!!!!!!!!"
I have no problem with that. He is trying to help me and I didn't follow through. I thought at first he wanted me to check them in circuit, then lift them and check them  again. I read the first part, which I thought was a separate step, because it was followed with "Now, onto the amp" which at first I thought was the next step. I kind of read them separately as two different steps. I didn't take it as a whole. When I checked them in circuit and one came up 0.0, I stopped there before going to what I though was the next step to ask for advice that was already given.I have been married to the same women for forty seven years I am used to such statements lol.  When working with a person who I know is way beyond me I try and shut up and follow instructions, I don't always get it right, and for that I apologize. I have read tons of books on the subject, can basically read schematics, know mostly what the individual components do and how they work together and how the different sections of the amp work together. It is the hands on that I have very little experience with. You can only learn so much from books. When I am teaching a novice guitar player, I realize how much is done automatically without even thinking about it and how basic it seems. Same with CNC programming and machining. When I play with a guitar player who is beyond me it is time for me to shut up and listen. I spent all this money on test equipment, scopes, meters, variax, isolation transformer, ESR tester, cap testers,  transistor checkers, tube checker, I certainly want to learn and I don't wan't to antagonize anyone trying to teach me.  I am at the age where soon I probably won't be able to play anymore, my hands are already giving me trouble. When that time comes, I would like to keep my hands in the business. Also, at this age I am not as sharp as I used to be and sometimes that can be very frustrating.Once again
Thanx all for bearing with me.
J.D.

[dragonworks]

Diodes came in and have been replaced. I built LBL and plugged her into that. No shorts, no fuses blowing. No sound. Won't come out of standby. All tubes filaments are lighting. All filaments are getting the proper voltage. In Standby mode I have about 450 volts to the plates of the 6L6s. In the middle position I have no voltage, in the bottom position I have the 450 volts again.  No matter which position I put the three way switch in there is no voltage to any of the plates of any of the other tubes. The switch has been rewired by I don't by know who. I am going to have to work on that next, see whats what.Thanx again.J.D.

[John]

I've been having a bit of tequila, so don't take this as gospel. But, I suspect your "standby" switch is either not wired right, or not functional. See if you can clip in a lead from the  second or third filter cap to the plates of your phase inverter. If you have voltage there (~300?) that's part of the "divide and conquer" thing done. And oh Yeah, Sluckey is the king of trouble shooting!

[PRR]

> In Standby mode I have about 450 volts to the plates of the 6L6s. In the middle position I have no voltage, in the bottom position I have the 450 volts again.

Are the marks rubbed-off on that switch? Center is "Standby", mute, no voltage. One way or the other way is Play, signal and voltage to all tubes, but buzz-reduction ("death") cap to one side or the other of the wall-cord.

Why the 6L6es get juice, but not other tubes, is a different and probably logical problem.

[Keppy]

The switch is not your problem. Your power tubes have plate voltage, so you know that the switch works correctly, and that you need to check the next couple of B+ nodes. Do you have voltage on the screens of the power tubes? What about after the 2.2k dropping resistor? (point F on the schematic)


https://el34world.com/charts/Schematics/files/Ampeg/Ampeg_g15_gemini_ii.pdf

[dragonworks]

Thanx for the help. I will be getting back to it soon. Started a new job, have to go to work, boo. Would be nice to learn to be an amp tech.

[dragonworks]

Forgot to mention. I found another schematic in my Jack Darr book from circa '65, that is much simpler to follow.  I will compare the two when I get back downstairs. I'll see if I can get a picture of it and post it.

[dragonworks]

So far. Using the schematic in the pictures. There appears to be no voltage coming out of the B+ C&E branch.

The 6L6s are getting plate voltage 500 volts, and -58 bias voltage

Tube V7 (12ax7) pin 1 fed by E+is getting 0 voltage, pin 6 is getting 150 V fed by F+.
Tubes V4&V5 (12ax7s preamps) show 0 voltage on all plates, all plates fed by E+.

Tube V6 (6CG7) pin 1 fed by C+ is showing no voltage, pin 6, fed by F+ shows 297 Volts.

Tube V3 (7199 phase inverter) pins 1+2, plates fed by C+ show +.2 and -.2. (vitrtually 0?)

At the node of the 15k 1W resistor and the 40uf filter I show 0 voltage.
At the node of the 5.6k 1W resistor and the  40uf filter I show 0 voltage.
All other nodes are showing  between 450-515 volts.
I am still at it. Now, to figure out why. Wait, wait, don't tell me. Yet. You guys probably get it right away?

[dragonworks]

I lifted the 510k resistor connected to the switch and the pilot light, it checks good. I replaced both 40uf filters with 47uf @ 450V, still no go. I checked the caps I pulled, they check in spec and good on my ESR meter. One note, when I have power to the plates of the 6L6s etc, (with the switch up), I show 0 voltage at the junction of the 510k resistor and the pilot light, and the lamp is off. When switched to the center position, I lose the plate voltages of the 6L6s etc, but get -440V at the junction of the 510k resistor and  the pilot light, and the lamp lights. The lamp only lights in the center position. When switched to the down position I get the plates back on line but lose the voltage at the 510k resistor and the pilot light and the lamp goes off. The lamp is lit and voltage applied to the junction of the 510k, and lamp  only when in the center position. Even with voltage at the junction of the 510k,  and lamp,  I still have no voltage to the plates of the preamps. I haven't tried lifting the 15k and 5.6k 1 watt resistors and checking them. If I do, and they check good, that makes me suspect the switch again? Am I on the right track? I am going to put the two caps back that I pulled, they are rated 39uf at 500V. The ones I replaced them with have a  voltage rating  50 volts lower than the callout on the schematic.

[dragonworks]

To answer another question, On the power tubes, on pin 3 plates, I have 500V, on pin 5 Grid 1 I have -58V, on pin 4 grid two I have 508V, on pin 8, cathode I show 0 volt DC.
The B+ F branch has power on two different plates, V6 pin 1, 297V and V7 pin6 150V.

[sluckey]

Quit mucking around with the switch, pilot lamp, and 510K resistor. Nothing wrong there.

[Keppy]

At the node of the 15k 1W resistor and the 40uf filter I show 0 voltage.
At the node of the 5.6k 1W resistor and the  40uf filter I show 0 voltage.
All other nodes are showing  between 450-515 volts.

Sounds like you have reasonable voltage at F, but not at C. Those points are connected through the 5.6k resistor. If C was shorted to ground, the resistor would pass 450v/5600Ω ≈ 80mA of current, for 80mA x 450v ≈ 36W dissipation. In other words, it would burn up. If the 5.6k is not burnt, then it's probably open rather than shorted. Meter that resistor to make sure the resistance is about right, and check it for continuity to the filter caps at both F and C.


Regarding the lamp, notice in the schematic that it connects through a resistor to point C. You can't expect it to light if C is at 0v.

[dragonworks]

You are talking to a novice here. I have no voltage at E either. My next step was to lift both 1W resistors and check them, they are not burnt. Next time I am downstairs I will get to it. I have to practice some PSG in the meantime. I am also going to do what sluckey said and quit mucking with the switch, lamp and 510k resistor.
Thank You

[Willabe]

He's saying you have B+ at the 1st B+ power node, now start looking down stream to find where the B+ stops flowing.

[dragonworks]

The 5.6k resistor is open and needs to be replaced. I have none at the moment.I lifted one leg and measured across and get no reading anywhere in the ohms range, used two different meters to be sure.
Should I check the 15k 1W next to it that it is connected to?I should order both, I have neither and I have a lot of resistors.Thanx again.J.D.

[Willabe]

Should I check the 15k 1W next to it that it is connected to?

Yes.

If you plan on doing more tube amp work, then with inexpensive parts, like resistors, when you place an order, always get a few extra to have on hand. Shipping charges get's ya these days for just a single resistor. 

[dragonworks]

I ordered twenty of each from Newark, 18 and change including shipping. I will go check the 15k and let ya know.

[dragonworks]

The 15k checks out good. I have two Magnatone M15s that need work and a Peavey classic 30 that needs work. Also an Ampeg Echo Twin that has about half of it missing. That is the last one on my list if I ever get that far. I have learned quite a bit through this thanks mostly to you guys. The classic thirty has either a bad socket or a bad solder joint on one of the EL84s heater pins. Tube 3 in the row goes out, all four go out. Reach in, wiggle it around and you are good for awhile. She has started squealing also. I ordered four new sockets and replacement caps for all the electrolytics.  I had it apart once for the heater problem, could find nothing wrong, cleaned and tightened the socket up and put it back together. Worked for awhile but started again. It is pretty old, I would think around ninety three. That will be the next one I get to if I ever get through this one.

[dragonworks]

The amp is now working. It does have a bit extra hum, I think it is 120 cycle. The tremolo is working. I haven't put it back in the cabinet yet to see if the reverb is working. I need to see if I can quiet her down some. Haven't messed with the hum balance, will try that first. I should switch that cap back in also.
Thanx to you guts it is now not just gathering rust and dust in the corner.On to the Magnatone M15 with a better understanding.In the meantime, need to practice my PSG.
Thank You all again.J.D.

[dragonworks]

The reverb is not working. The tank is good, I can hear the thunder when I tap it. I suspect no signal is getting to it. I will have to look into that. The hum balance took some of the hum out but there is still way too much.