Powering trouble with the 5672

[~arph]

Hi,

I"m facing a strange issue. I'm working on a compressor based on 5672 tube.
Basically I build a pentaboost as a basis, see:

http://www.diystompboxes.com/smfforum/index.php?topic=70524.0

Currently I have the pentaboost on the breadboard and it works.
The problem I have is related to the power supply.

When I use the 1spot. as the power source for the schematic above. The V+ drops to around 3.6 volts. If I disconnect the heater voltage (the 1.25V output from the 317) the V+ is back at 9V.
When I use a battery, this issue does not appear (which will explain my noted gain difference between 1spot and battery as I posted before)

To me this looks like a current draw problem around the 317, but maybe someone here has different ideas.
Perhaps the way the LM317 is used here is not correct as 1.25 is the absolute minimum it can put out. If so, is there a different way to get a reliable 1.25V heater voltage, disregardless of the actual V+ (say 9V and up)

Thanks,

Arnoud

[frequencycentral]

I have had no problems using LM317T.

There is a thread about using a voltage drop resistor instead of the LM317, but the resistor value is dependant  on the power supply voltage: http://www.diystompboxes.com/smfforum/index.php?topic=71414.msg576012#msg576012

Jasper Ooosthoek over at AX84 who redesigned the Red Star Drive had this to say  about using the LM317:

Jasper: "I took the liberty to remove the 100 ohms resistor from your original circuit. It's true that the LM317 outputs 7 volts (I measured it) without any load. But the tube filament is a 25 ohms resistor. Therefore it is low enough to ensure the voltage is not larger than 1.25 volts. Anyway, my circuit worked brilliantly without the resistor."

Rick: "I'm not quite understanding the LM317 minus the 100 ohm resistor thing. Logic tells me the heater is getting too much juice - but there are a lot of things in this world I don't understand."

Jasper: "You need 100 ohms or smaller resistor to prevent the voltage from rising above 1.25 volts. This means a minimum current of 12.5mA (1.25v/100ohms) has to flow. The heater is a 25ohms resistor (1.25v/0.050A). Which is much smaller than 100 ohms, also the current of 50mA is four times larger than the minimum current that has to flow."

I don't get it - but it appears to work for him!

The compressor idea is cool by the way! 

[~arph]

Interesting, but the funny thing is that with a battery I don't have this problem.. It only appears with the switching powersupply (1spot).
I'll experiment more with the LM317. Meanwhile I'm going to experiment with the two diodes option..

Regards,

Arnoud

[frequencycentral]

..........or, just for now, you could use a 1.5 volt battery for the heater. That's what Renegadrian did with his Pentaboost. 

[~arph]

Yes, that is always an option, but the least desirable. I'm just keen on finding if there is a logical explanation for my observations.

Tonight, I will try it without the 100R at tpin 2 of the 317 and the second thing I'll try is running the heater at 0,7V as this should work too accoring to the original poster of the VCF (on the dutch synth forum)  Maybe the last option requires me to drop the 9v a bit too, but I can do that with a zener. to say 7.6 or 6.2 v as long as I can still power a dual opamp on the same supply. Slightly above unity gain is good enough for me in this application.

[frequencycentral]

Yeah, I still didn't build that VCF yet, although that's the reason I bought my first batch of 5672. I think the BC547 in the signal path put me off. Also, from what I have read, running heaters below their requirement can cause cathode stripping. Whatever resistor value you use, make sure it is 1 watt rated though.

What LM317 type are you using?

Maybe it's the One Spot - isn't that a switched mode supply? Maybe the tube doesn't like switched mode.

[~arph]

Yes, the author talks about the cathode stripping.. he's been running some 5672's at 0,7 heater voltage for over 15 years now,  never had any trouble..
I have the LM317 not the LM317T. I have to check this now as I just read on the site I bought it from both T and not T in the description... :S

Yes the 1spot is a switching powersupply, it does appear to be of influence since on a regular 9V battery the issue is nonexistent.