DC Filament hookup

[liddokun]

Hi folks,

I know that the topic of should you or should you not use DC heaters has been discussed many times before, this is not that.

I am going to be using DC filament on my preamp section for two amplifiers.

I am using a pre-made regulated+filtered pcb from Pete Millett. The schematic is attached.

I will have a separate transformer feeding the board.

Everything is built up, but I'm a little confused as to how it gets hooked up. I'll be running this at 6.3volts. As I have this now, I have pins 4 and 5 tied together and going to the + output of the DC board. I am a little unsure what to do with pin 9. With 4+5 connected to the positive output of the DC board, and the negative output going to chassis ground, I get a stable voltage output which I'm able to adjust to 6.3 with no issues.

With pin 9 connected to the negative output of the DC board, the voltage regulator heats up very quickly and I assume is going into thermal shutdown as the voltage output rapidly drops to about 1v.

A little guidance would be helpful.

[Stevem]

How many tubes are you powering off of that 1084 regulator?

Do you have a heat sink on it and in fact a dam big one.

At work where I handle the repairs we sell a audio compressor that uses this regulator and it powers the filaments on 3 preamp tubes and a 6AQ5 output tube and the 1084 is bolted down to the units chassis, so it’s using the units whole chassis as a heat sink.

Even still if you touch the chassis right where it’s mounted it will burn your finger in 10 seconds or so.

If you have it mounted to one of those stand up type heat sinks that are not much bigger then the regulator itself then I don’t think you have enough mass there to stop that 1084 from going into thermal shutdown.

[liddokun]

I'm powering 3 12ax7s.

I've got it bolted to the amp's chassis. The chassis gets mad hot as you've described as well.

I'm considering switching the transformer and running the heaters in series at 12.6v, but I did want to make sure I'm not just hooking this up wrong.

[sluckey]

I've got it bolted to the amp's chassis. The chassis gets mad hot as you've described as well.

The tab is connected to the output! You must use an insulator kit if you will connect to chassis.

[liddokun]

Hey Sluckey,

Yes, I've got it insulated! (learned this the hard way in the past)

[Stevem]

Do you have C3 4 and 7s polarity correct?

Are they good?

If you lift them one at a time does the voltage output level return?

[sluckey]

Hey Sluckey,

Yes, I've got it insulated! (learned this the hard way in the past)

Unbolt it from the chassis and just let it float in the air. Does it still get hot?

[nuke]

Look for a short in the wiring, and then try it with the tubes out of the sockets.

Is one of the 12A*7 a cathode follower stage?

[liddokun]

Do you have C3 4 and 7s polarity correct?

Are they good?

If you lift them one at a time does the voltage output level return?

The polarity is correct. The circuit works as it should when not hooked up the tubes; ie outputs the appropriate voltage

Hey Sluckey,

Yes, I've got it insulated! (learned this the hard way in the past)

Unbolt it from the chassis and just let it float in the air. Does it still get hot?

Yes, it does.

Look for a short in the wiring, and then try it with the tubes out of the sockets.

Is one of the 12A*7 a cathode follower stage?

No shorts that I can see. I'll give it a shot with no tubes.

Yes, JCM800 circuit so one half of V2 is a cathode follower.

[nuke]

Cathode followers can be problematic, due to heater-cathode leakage or perhaps even a short.

[B Ingram]

I'm powering 3 12ax7s.

I've got it bolted to the amp's chassis. The chassis gets mad hot as you've described as well.

The diagram includes the recommended heatsink for your regulator.

[liddokun]

I'm powering 3 12ax7s.

I've got it bolted to the amp's chassis. The chassis gets mad hot as you've described as well.

The diagram includes the recommended heatsink for your regulator.

I assumed bolting to the chassis would be a far better heatsink as it's much larger than that heatsink.

My assumption is that, as there is continuity between pin 09 and pin 4+5, once I connect the positive voltage to pins 4+5 and pin 9 to ground, the regulator is just seeing it's output being shorted to ground, causing excess heat and thermal shutdown

[Stevem]

Here’s how a LT 1085 is used ( same family of regulators ) in the gear I posted about.

Is your insulation washer good when it’s under the hold down machine screw and things are clamped down?

If so then I just plain think you need a new 1084.

Yes, bolting it to chassis without a short taking place to the tab is the only way to go.

[liddokun]

Based on the schematic, it would appear I am hooking everything up correctly, but seems that even bolting to the chassis is not enough heat dissipation to prevent thermal shutdown.

I've tried with several 1084s and 1085s with the same behaviour.

I'm using silicon insulator sheet designed for to-220; as well as nylon screw and nut to prevent shorting.

[martin manning]

I'm wondering if the low resistance of the cold heaters is pulling so much current that the regulator just can't recover. You might try putting a resistor across the output that is equal to the normal heater current and see if it can handle that. Each 12AX7 with heaters wired for 6.3V will have 6.3/0.3 = 21 ohms resistance once it's warmed up.