"Fiddy" 50-watt build log

[xtian]

Taylor, yes, this circuit is the 1987, and it has a cathode follower. All stock values.

When planning, reading, dreaming about this build, I was trying to absorb all the excellent choices for mods to the RC pairs for the preamp stages. Everyone (Reeltarded!) has their favorites, and it quickly became overwhelming. I even planned ahead for a cathode bypass switch, but decided to start with stock values, and ended up not bothering to wire up the switch. The end result, using the stock values, is wonderfully variable in tone by adjusting the Volume 1 and Volume 2 controls, and I have no urge to tinker!

Gonna replace the burnt fuse today and hope for the best.

[xtian]

Experiment #2 = FAIL!

I got new fuses. I put in a new faux heater CT using 150R pair in a different location, off of V3. This time when I powered on (standby off, so no HT yet), I could hear a hum through the speaker that was not there before. And when I flipped on the standby switch, the two 150R resistors burned out immediately, with a mighty blue flash. Fuse did not blow this time.

So I pulled all the tubes, pulled the burnt resistors and disconnected the heater pair from the sockets, so the PT is no longer connected to the heater chain, but all the sockets are still connected together. I measured conductivity, and the heater chain looks perfect. No shorts to ground or cross shorts, and good conductivity along each run.

Hooked the PT's heater pair back up to the sockets and powered on (still with no tubes) and all PT voltages look fine, including the heaters.

So, WTF is causing a huge current spike thru the heaters and into the chassis ground through the artificial CT?

Again, I'm wondering, do I need the CT? Isn't it just a solution for hum, or do I really need the reference to ground?

[xtian]

Well, let me answer my own questions, after more reading at Valve Wizard. I surmise the artificial CT is used to reduce hum, but is not essential to operation. So for trial #3, I left the CT out.

I replaced the Mesa EL34s with my only other pair, a mismatched poo pile. And the amp functions! Though the bias 1R resistors read 18 and 40mv…a little *ahem* unbalanced.

So: BAD TUBES! Why would a bad tube cause this ugly short through the heaters?

OK, Steve, I'll be ordering new tubes now...

[Colossal]

Aaron,

Quick question: can you take the OT out of the equation? My guess, taking a cue from Martin's idea, is that the OT took a shot. If you can just float the amp to B+ and it stays on, then you might be able to decide if the OT is toast or, if it pops again, then there is a wiring issue somewhere else that is just sinking current till the fuse or weakest link (100R pair) blows. You do not have to use the CT, even if the PT has one.

[ToneMerc]

Experiment #2 = FAIL!

I got new fuses. I put in a new faux heater CT using 150R pair in a different location, off of V3. This time when I powered on (standby off, so no HT yet), I could hear a hum through the speaker that was not there before. And when I flipped on the standby switch, the two 150R resistors burned out immediately, with a mighty blue flash. Fuse did not blow this time.

So I pulled all the tubes, pulled the burnt resistors and disconnected the heater pair from the sockets, so the PT is no longer connected to the heater chain, but all the sockets are still connected together. I measured conductivity, and the heater chain looks perfect. No shorts to ground or cross shorts, and good conductivity along each run.

Hooked the PT's heater pair back up to the sockets and powered on (still with no tubes) and all PT voltages look fine, including the heaters.

So, WTF is causing a huge current spike thru the heaters and into the chassis ground through the artificial CT?

Again, I'm wondering, do I need the CT? Isn't it just a solution for hum, or do I really need the reference to ground?

Don't worry about the artifical CT, it's actually saving your hide. Start dividing the PS circuit up and troubleshoot it in segments. The easiest way to shotgun it would be it start disconnecting load segments until the short dissappears.

I would do this first, disconnect the output transformer primary


TM

[xtian]

Thanks, Colossal. As shown with my last test, the amp is functional with poorly matched tubes. So the OT is working, at least nominally.

Guess what I found during post mortem? One of the Mesa EL34s reads 4 ohms between pins 2 (heater) and 3 (plate). Ahem. Aiee!

So let's revisit the original flameout: We're cranking things up (though disaster is nigh because I've plugged in my 2ohm OT into a 16 ohm cab) when suddenly, the heater element in one of the power tubes shorts to its plate, sending loads of current thru the heaters into the ground and burning out the fauxCT, then, mercifully, popping the HT fuze.

Let's see, who was it said, "Bad tube"? Scrolling back… WINNERS: ToneMerc (first mention), and Cameron (more declarative mention)!!

[martin manning]

Well, let me answer my own questions, after more reading at Valve Wizard. I surmise the artificial CT is used to reduce hum, but is not essential to operation. So for trial #3, I left the CT out.

I replaced the Mesa EL34s with my only other pair, a mismatched poo pile. And the amp functions! Though the bias 1R resistors read 18 and 40mv…a little *ahem* unbalanced.

So: BAD TUBES! Why would a bad tube cause this ugly short through the heaters?

OK, Steve, I'll be ordering new tubes now...

If it functions with another set of tubes then I think you have found the problem that took out the second HT fuse- you have an internal short in an output tube. The first failure could have been caused by a shorted tube, or high flyback voltage from the load mismatch could have caused multiple failures including a tube. I'd do a little checking on the OT to see if the windings have reasonable DC resiatances, and maybe run a check of the voltage ratios from each side of the primary to the secondary with an qapplied AC voltage.

[xtian]

Martin, how much of the OT do I need to disconnect in order to test, and what should I be looking for?

[martin manning]

Martin, how much of the OT do I need to disconnect in order to test, and what should I be looking for?

Just disconnect the CT, and the NFB too if you like. With the tubes out, put a 1/4" plug in the speaker jack. Then with the power on and standby off, connect a clip lead from the filament voltage to the tip of the speaker jack. That should put ~3.2VAC on the OT secondary. Measure that voltage and also measure the voltage on each side of the OT primary (CT to plates) and across the plates. You should see the same voltage on the two sides. Calculate the ratio of the plate-to-plate voltage to the secondary voltage and square it, then multiply that result by the selected speaker tap ohms and you should get something close to the right primary impedance.

[xtian]

Thank you, Martin! Here is my math:

connect a clip lead from the filament voltage to the tip of the speaker jack. That should put ~3.2VAC on the OT secondary

My heater supply currently has no CT, so I had to use TWO jumpers to clip heater voltage to the OT primary. I read 6.73vac.

Plates to CT read 149.5vac both sides, 299vac across plates. 299/6.73 squared, multiplied by 2ohms (the only output tap) = 3948. The spec'd primary impedance is 4150, so that sounds good!

Really appreciate the detailed instructions!

[martin manning]

And that 6.73VAC was with the filament winding connected to the OT secondary, right?

Looks like the OT is not fried, but it's still possible that it has sustained some damage that will only show up with the full prinary voltage applied. I'd reconnect the heater CT because you'll need it to keep 60Hz hum down. Look over your heater wiring for damage and keep the leads away from the plate lugs on the output tube sockets. In the crop below, it looks like you have the heater leads running over the plate lug with the brown OT lead connected to it. Look for evidence of arcing there, and I think I'd reroute those leads.

[xtian]

Yes, that 6.73vac was with the filament winding (both legs) connected to the OT secondary.

I did indeed reconnect 100R resistors as an artificial CT last night, and successfully booted up and played for over an hour, quietly, with no trouble. Baby is still quiet as a mouse in terms of hum and hiss.

I examined everything carefully; the heater chain is intact and healthy, and there never was arcing, except inside the bad EL34. The only fireworks were the 100R resistors burning up like fuses.

Thanks again, Martin!

I'm sending of the faceplate artwork today to Kyle at panelcraftinc.com.

[Milkmansound]

nice! I just started working with Kyle - they are on point! Can't say enough good things about them

[xtian]

Faceplates are in. Awe-freaking-some.

[Structo]

I am feeling kind of dense about this.

I understand elevating the heater voltage above ground to reduce hum but I have never heard of connecting the heater center tap or artificial center tap to the output transformer secondary.

What is the theory behind that?

The way I have seen it is to connect the CT to a cathode or other low voltage source to elevate the 0 voltage above ground.

Care to 'splain that to a dummy?