DIY 1624 from Hammond M3 Parts

[Smokebreak]

That's a good question, and I'm unsure. I added R95-R99, as that string seems to be a voltage divider, in addition to providing some lower voltages to other parts of teh M3, that you don't need. Thats how I came to ~10K

For instance, it's easy to calculate the power for R100. 20V dropped across 2K2 gives 9.09A of current, using Ohm's Law. The Power dissipated would be voltage x current = .18W . Push that up to 1W to be on the super safe side.

That 10K though, looks like it will drop the full 310V to ground, on power down, which would make it something like a 20W bleeder resistor. So, I'm unsure. Honestlly, I'm unsure if that 10K even needs to be in there to get you to your target B+

[dawsonaudio]

It's been a couple weeks since my last post here....finally have some time to finish my amp build.

Referring to the schematic, I'm assuming I can jumper a wire across the 6973's to add B+ to both pins (1 & . Or should I run individual wires to each pin from my power supply? I'm assuming I need to connect both pins 1 & 8 together, as well as 3 & 6 together. Is that correct?

[img2100]http://www.dawsonaudio.com/1624T.jpg[/img]

Thank you for the help,
Nate

[dawsonaudio]

Actually, as I look inside the 6973's that I have, pin 2 and pin 8 don't connect to anything inside the tube. They dead end once they enter the glass. All other pins connect up.

[dawsonaudio]

I'm wondering if the tubes that I bought are really 6CZ5's relabeled to be 6973's. I notice that they are almost identical. The 6973's have connections to both pin 1 and 8, but the 6CZ5's only connect at 1 and not at 8. Both tubes have pin 2 not connected.

What was the inverse of the Westinghouse slogan? "You can't be sure if it's a Westinghouse?"

[Smokebreak]

Correct. If they aren't true 6973s looks like they will need connection only to pin 1 for screen connection. If they are 6973, then 1,8 would be connected internally, so you wouldn't need the jumper anyways . Generally speaking, if say you did need a connection from one point to 2 other points, use the least amount of wire you can to accomplish the connections.
Pins 3 and 6, the control grid, should be internally jumpered as well. You can always test this with your meter for continuity as well.

[dawsonaudio]

Yeah, pin 3 and 6 are connected internally and show continuity. I notice that the 6CZ5 is similar, but I was told these were 6973 tubes. Maybe I got taken for the novice I am with vacuum tubes.

[Smokebreak]

I wouldn't sweat it too much. They seem to be nearly identical tubes, except for the pinout thing, and the fact that the 6CZ5 has lower max values for plates and screens. When you get to the fire up process, observe the tubes for redplating. Folks exceed the "max values", and get away with it sometimes, you're just gonna have to see.
If they don't hold up, you can always get some new production 6973s, or some $NOS$, or lower that B+ if need be.

[dawsonaudio]

On that revised schematic of the power supply section that you worked on, does the negative side of the first 20uf tie in at the #4/100v test point and feed through the field coil to ground? I'm assuming that is what is happening.

Nate

[Smokebreak]

On that revised schematic of the power supply section that you worked on, does the negative side of the first 20uf tie in at the #4/100v test point and feed through the field coil to ground? I'm assuming that is what is happening.

Nate

I'm not exactly sure what you mean by "feed through"...

The PT centertap, the negative side of the 20uf, and one side of the field coil should be connected.
The other side of the FC is ground. This is what is generating the neg voltage, and for your purposes, the relative B+ voltage drop..hopefully

[dawsonaudio]

That is what I was trying to ask. The other end of the field coil ultimately connects to ground. Just wanted to make sure of that. I will wrap things up this evening and hopefully post some images tomorrow...if I don't run into any problems.

Thanks again for all the assistance here.

[Smokebreak]

Here is a good "first fire up" guide you can follow : http://www.paulrubyamps.com/info.html#FirstPowerUp

I would also go over the schematic and your wiring a couple times, to insure everything is connected properly, especially the power supply.

You may also use a light bulb limiter to prevent any catastrophic losses in case of faulty wiring : https://tubeamparchive.com/viewtopic.php?t=20341

Good luck!

[dawsonaudio]

Here are a couple images of the amp before I blow it and myself up upon first start-up. I will post some more once I test it and mount it in my cabinet later this evening.

[img600]http://www.dawsonaudio.com/Top.jpg[/img]

[img600]http://www.dawsonaudio.com/Under.jpg[/img]

[Smokebreak]

That is quite the transformation! Great work!

I see you used ceramics for the couplers like the originals. Bonus points.

[dawsonaudio]

So I just started testing my amp. A couple things are happening...the 10k/10 watt resistor tied in at the first power supply cap b+ to ground is getting really hot. I assume it is suppose to get hot. So I got 430 volts d/c coming off the first two caps in the power supply and 420 at the final cap in the supply. B+ was also present at the OT and at the 6973 power tube sockets at pin 9, connected to the output transformer, and pin 1, connected to the second power supply cap b+. This was all checked with just the rectifier tube inserted, 5U4. The input voltage off the transformer to the 5U4 is 370volts a/c.

I went ahead and installed all the signal/power tubes and powered up. I noticed one of the power tube center not lighting up, while the other power tube lit up quickly and started to glow blue on the very bottom. So I turned off the amp.

Also, I swapped the power tubes to rule out a socket issue, turned the amp on quickly and the problem followed the power tube and not the socket. Turned off the amp. How should I go about checking the power tube that isn't lighting up?

Should that 10k/10 watt resistor be that hot? I'm noticing a little light smoke rising from it. Looks like it's turning black in the middle.

Thanks.

[martin manning]

Power = VxV/R, 430x430/10000 = 18.5W