Help with Ampeg B15n project

[bubbab_74]

I'm building a B15n based off of the Joe Piazza schemo ('68 model) with the following exceptions:
1 channel
60W Hammond 1650P OT
Hammond 274BX PT

I keep blowing the fuse. Using a 3A/250V slo-blo. With power on, it's fine, but as soon as I flip standby on, it fries the fuse. With standby off, I get voltage readings of 415vac on each secondary and around 190vdc on pin 8 of the GZ34. Heaters are reading 5.4v on the 5v heaters and 7.1v on the 6.3v heaters (expected with the Hammond trafos being rated at 115vac and bringing in an actual 121vac). Also, if the standby is switched on with no rectifier in place, it's fine and the neg. bias supply reads about -51vdc.

I can't find any shorts. Ideas?

[throbbinnut]

Something is either wired wrong, or shorted to ground. Can't tell you much more than that without being there to look at it.

If you don't have a variac so you can bring it up slowly, try using the 100W light bulb trick (google it - you wire up a 100W 120V bulb in series with the amp, if the amp is working normally, the light bulb may barely glow, if the amp is drawing heavy current, the light bulb will limit it to about 1A and save the fuse from blowing) and you may be able to get it powered up enough to see where your current is going before the fuse pops.

Chris

[aborgman]

Originally Posted by bubbab_74

I'm building a B15n based off of the Joe Piazza schemo ('68 model) with the following exceptions:
1 channel
60W Hammond 1650P OT
Hammond 274BX PT

I keep blowing the fuse. Using a 3A/250V slo-blo. With power on, it's fine, but as soon as I flip standby on, it fries the fuse. With standby off, I get voltage readings of 415vac on each secondary and around 190vdc on pin 8 of the GZ34. Heaters are reading 5.4v on the 5v heaters and 7.1v on the 6.3v heaters (expected with the Hammond trafos being rated at 115vac and bringing in an actual 121vac). Also, if the standby is switched on with no rectifier in place, it's fine and the neg. bias supply reads about -51vdc.

I can't find any shorts. Ideas?

All the standby switch is doing is floating the PT center tap/attaching it to ground. I can't imagine why floating it (standby on) would cause the fuse to blow, unless the standby switch itself had something wrong with it.

[beans-on-toast]

Good advice from Chris using the light bulb limiter. There is a short in there somewhere.

You have to break the problem down and look at each part in turn.

First of all, if you haven't already done so, the 0.05uF 1000V cap and switch on the primary side of the power transformer should be removed. This is what is called the death cap and is unsafe.

Your power cable should be three conductor with the safety ground wire connected to the chassis at a location away from any other grounds.

The newer version of the Hammond 274BX has both 115 (GRY) and 125V (BLK) taps. If you have this version, use the 125V tap. Wrap the end of the unused tap with an insulating tape and tuck it away.

Check all the wiring. If you are following the schematic, the center taps (YEL/BLK and GRN/YEL) from the two heaters should not be connected. Make sure that they are ISOLATED and tucked away as above. Do not short them together. In the case of the 6.3 VAC tube heater circuit, you use either the balance pot, two resistors, or the heater center tap. Use only one.

Ideally, you want to test the secondaries of the power transformer with nothing connected to them and power connected to the primary side. It is a good idea to do a no load test before the transformer is wired in to check it out.

Remove all the tubes. Check the voltages of the high voltage, 5 V and 6.3 V windings. Both with the standby on and off. Check that the high voltage center tap circuit with the standby switch has no stray wire strands coming into contact with anything it shouldn't.

[simojo]

I had a similar problem with a build. There certainly was a short to ground, on the rail, which happened as soon as the standby was flipped. In my case, the board was fairly close the chassis, and a wire or lead that I soldered on to it was tiny bit too long. I finally used a dental mirror and flashlight to look under the board, then I saw the short, where the long wire reached all the way to the chassis.

[BassmanPaul]

When building an amp it's best if you build and test in stages especially for someone new. Build the power supply first and then test it. Once you have verified its operation you can move on to the power amp stage. Again you can test this and if correct you'll have the pleasure of some signal appearing in the speakers. Finally build the pre-amp. Once checked you're done

Remember that the feedback loop has to be negative and not positive. If when you are testing the output stage it oscillates, reverse the connections to the plates from the output transformer primary.

[bubbab_74]

Ok, so I put it on the light bulb limiter, checked all the wiring, didn't find a short, but rewired a couple things anyway. Turned it on and no pop! However, my secondary coming into the filter caps is over 550v and the first cap is an F&T 30u/500v, so I'm going to have to switch that out. I'll come back when it's all taken care of. Thanks so much for the help, guys.

[aborgman]

Originally Posted by bubbab_74

Ok, so I put it on the light bulb limiter, checked all the wiring, didn't find a short, but rewired a couple things anyway. Turned it on and no pop! However, my secondary coming into the filter caps is over 550v and the first cap is an F&T 30u/500v, so I'm going to have to switch that out. I'll come back when it's all taken care of. Thanks so much for the help, guys.

Is that with tubes installed, or not?

Without tubes installed the lower current draw through the PI filter network/rectifier won't drop the B+ as much as it would when drawing a normal idle current - so it may not be that far off.

[BassmanPaul]

For the primary filter it's always best to double the capacitors in series with a 220KΩ 2W resistor across each to balance the voltage across each. Look at Fender amps for an example of this.

[simojo]

Glad you "fixed" the short! I believe that the incoming surge at turn-on can spike the voltage much higher than the idling voltage. Even if you steady-state idle at 500V, you can pop a 500V cap eventually. This is why Fender and Ampeg did this, as Paul pointed out. I talked with Dennis Kager this summer at his shop. He mentioned that he made this change on the B15 specifically for this reason. Compare the B15NC schematic vs the B15NF, you'll see how he upped the first cap in 1968. Smart guy.

[bubbab_74]

Here's my schematic. You'll see I replaced the 30u/600v filter cap with two 100u/350v caps in the PS. All previous tests were without tubes except for the rectifier. I ran a test after no fuse blowing with all the tubes in and I got signal through, but as soon as I turned it up a bit and started playing, it blew again. I swear I cannot find a short anywhere unless there's something wrong with the schematic. Take a look and tell me if something's odd or wrong. If anyone wants gut shots, I'll take some tomorrow and post whatever areas you'd wanna look at. Thanks.

[bubbab_74]

Originally Posted by beans-on-toast

First of all, if you haven't already done so, the 0.05uF 1000V cap and switch on the primary side of the power transformer should be removed. This is what is called the death cap and is unsafe.

Your power cable should be three conductor with the safety ground wire connected to the chassis at a location away from any other grounds.

Never did install the death cap, just ran the common to ground as usual.

The newer version of the Hammond 274BX has both 115 (GRY) and 125V (BLK) taps. If you have this version, use the 125V tap. Wrap the end of the unused tap with an insulating tape and tuck it away.

This one only has the 115v primaries.

Check all the wiring. If you are following the schematic, the center taps (YEL/BLK and GRN/YEL) from the two heaters should not be connected. Make sure that they are ISOLATED and tucked away as above. Do not short them together. In the case of the 6.3 VAC tube heater circuit, you use either the balance pot, two resistors, or the heater center tap. Use only one.

Grounded the CT for the 6.3v heaters (GRN/YEL) and isolated the 5v CT (YEL/BLK).

Everything else looks fine, but I may be missing something.

[Vintage-Blue]

I can't help with anything technical but here is a copy of my B15N schematic in case it's useful to compare against the one you're working from. (the original was larger than my scanner so I had to scan it to two pages)

[throbbinnut]

You don't show a ground connection in your bias supply area. If the bias supply isn't referenced to ground, you're gonna get all sorts of freaky stuff happening. You'll have no bias - meaning your tubes are going to conduct like crazy until the fuse pops.

Chris

[throbbinnut]

If it ends up being the missing ground in the bias supply, you also need to confirm that the 100K dropping resistor to the HV tap you show is actually like 5 Watts rated. That thing actually dumps some power to get you down to 50VAC for bias from a 375VAC tap. If you want to be all ingenious, replace it with a .47uF 600V cap since then the power will be imaginary. Check the V4B schematic for an example.

Chris

[simojo]

Pics of the board may be good. Layout and lead dress are important. I've read about, but never actually done, this mistake: Lead dress is so poor that you have a high-freq oscillation in the circuit, which is ultrasonic but soaking up a lot of power. You then blow fuses at medium volume because of all the un-heard power being soaked up. It's a maybe.

The un-grounded bias supply is also very suspicious. Try that first.

[beans-on-toast]

In addition to what others have suggested, I have some concern about the inrush current with those large 50uF caps. You might be exceeding the peak inverse current of the 5AR4 tube. Look for a flash in the rectifier tube when you turn off the standby or before the fuse blows. If you see it, this isn't good.

I would try adding a diode in series with each high voltage wire from the transformer like in the image below. I use a 1N5408 (3A) diode but a 1N4007 (1A) will work. Try tacking the diodes in and see if it helps. For a permanent hookup you can connect the HV red wires to unused pins 5 and 7 on the tube rectifier socket, then run the diodes between pins 7 -> 6 and pins 5 -> 4. The leads on the 1N5408 are larger and are more difficult to work with.



From Fender 57 deluxe schematic.

[bubbab_74]

Thanks for pointing out the bias supply ground, it's actually a mistake in the schematic I forgot to change, but it is definitely connected in the amp and the tubes are getting their -50v. I'll take a couple pics and post them in a bit. It's all point-to-point and my lead dress could be better, but it's not terrible and I do plan on cleaning it up a bit before I finish.

Chris - Interesting idea with the cap on the bias supply. Any sonic benefits to doing this?

[throbbinnut]

You may have an instability problem since your driver and preamp both share the same power supply tap. If you look at the original schematic, the power supply for the driver is decoupled from the preamp power supply. You would need another cap and a dropping resistor to make a "D" power supply tap on your schematic. One way to see if that is what's happening is to pull the preamp tube V1, but leave the driver tube V2 in place along with the 6L6's and see if it will stay powered up that way.

Chris

[throbbinnut]

Oh, and you'll want to add some 220K 1W resistors in parallel with your stacked 350V caps. They make sure the voltage is divided equally across the capacitors.

And there's not really any sonic improvement by using a capacitor for the dropping element in the bias supply, it just runs cooler than a resistor.

Chris