Heater Elevation

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
16,155
Location
Norfolk - UK
I have been contacted several times over that last few months and asked to explain what elevated heaters are and why they are necessary so I thought it might be useful to post an outline description here.

Heater elevation becomes necessary because of certain circuit topologies and tube specs. Most tubes have a specification for the maximum voltage that is allowed between the heaters and the cathode. In many designs, like a conventional common cathode amplifier, where the heater is just a few volts above ground, this is not an issue, but in designs where a cathode can be 100V or more above ground, such as a cathode follower or a mu follower or an SRPP stage, it needs to be considered.

For a mu follower like the poor man's tube gain make up stage, or the SRPP output stage stage in the Eurochannel, where one tube sits on top of the other, the top cathode is at about half the supply voltage or around 140V above 0V. If the heaters were connected to 0V then they would be 140V below the voltage of the top cathode.  We also need to think about signal conditions. At the  maximum output of about 20V rms, the peak voltage on the top cathode would rise to 140V + 20*1.414 =  168V. This may  exceed the heater to cathode voltage spec of many tubes.  However, if we raise the heater voltage to about 80V above ground, then we reduce this to 88V and remain comfortably within the heater cathode voltage spec of most tubes.

The specifications for maximum heater to cathode voltage vary widely between tube types and even between versions of the same type of tube or between manufacturers of the same tube type. The maximum voltage the heaters can be above the cathode is often different to the maximum voltage the heaters can be below the cathode. Some double triodes even have different specs for each triode within the envelope.

The way heaters are normally elevated is by using a simple potential divider connected across the HT supply. The bottom leg of this pot divider is usually decoupled so the heaters at 0V for ac signals. There is often a specification for the maximum resistance between the heaters and the cathode so this needs to be taken into account in designing the pot divider.

I have attached a snippet from an RCA tube mixer power supply from the 1940s showing how they did it. The pot divider is 4R3 and 4R4 and the decoupling capacitor is 4C3. Notice the pot divider is connected to the hum dinger pot for the heaters.

Cheers

Ian
 

Attachments

  • RCAPSUheaterelevation.png
    RCAPSUheaterelevation.png
    271.1 KB · Views: 160
Thanks for the post.
Why does elevating the heaters reduce hum or buzz?
For the RCA circuit you posted, if the hum dinger pot started at ground and was increased to the V*18k/138k (with additional circuitry), where would the hum/buzz be minimal? Why?
 
dmp said:
Thanks for the post.
Why does elevating the heaters reduce hum or buzz?

I don't think I said that but in fact it often can. Some hum gets to the cathode when it is positive with respect to the heaters. The cathode acts like an anode and electrons can flow from the heaters to the cathode. If you elevate the heaters then the cathode is negative with respect to the heaters so there is no electron flow to the cathode from the heaters. This can reduce hum.

For the RCA circuit you posted, if the hum dinger pot started at ground and was increased to the V*18k/138k (with additional circuitry), where would the hum/buzz be minimal? Why?

I don't know. It probably depends on the particular circuit but I would guess it happens once you get  few volts above the cathode voltage. There is some evidence that elevating heaters makes tubes last longer.

Cheers

Ian
 
It's rare in old literature to see mention of elevated heaters doing anything other than reduce noise (when properly set).  It should be noted that RDH and many other sources also mention that hum is usually worse with DC heaters than with properly arranged AC heaters (bonus with DC elevated AC heaters); that being before the time we could swamp with multi-thousand mfd caps, but also an indication that DC heaters are not necessarily a savior, possibly just a sloppy design shortcut. 

Ian is certainly right in regards to heater/cathode difference, which was seldom an issue with old style circuits. 
 
DC heaters need a bridge rect, which can create junk that need filtering,

bypass the computer caps with a 0.1 uf


i have a tube rect on the phantom supply here, so no spike jones on the mic power,

silkscreen worn off, think it is a 6AL5 like you see in the old compressors

chews up a little voltage which is nice because phantom power transformers are usually oversized since the xfmr industry is un aware of our needs for 48 vdc.

 
emrr said:
It's rare in old literature to see mention of elevated heaters doing anything other than reduce noise (when properly set).  It should be noted that RDH and many other sources also mention that hum is usually worse with DC heaters than with properly arranged AC heaters (bonus with DC elevated AC heaters); that being before the time we could swamp with multi-thousand mfd caps, but also an indication that DC heaters are not necessarily a savior, possibly just a sloppy design shortcut. 

Ian is certainly right in regards to heater/cathode difference, which was seldom an issue with old style circuits.

Thanks for all of the info, but then why aren't the G9 heaters elevated, as the output configuration looks similar to the EZ?
Thanks in advance!
Best,
Bruno2000
 
G9 is a DC supply, and heater to cathode difference can be 200V with that tube. 
 
Back
Top