diyAudio (
-   Tubes / Valves (
-   -   Airgap for filament transformer ? (

Lauscher 27th April 2009 02:12 PM

Airgap for filament transformer ?
i have a SE amp with D3A interstage to 2A3. I use a separate filament transformer for the rectifier. 5V - 2A . First i have used 5R4GY, now GZ32 for slow start.
In use after 30 minutes, the filament transformer warms up a lot, near to hot. It is just the iron core. The coil is not hot . Dimensions of the EI-core: 60mm high x 55mm wide x 35mm deep.
The amp has 292V DC and 112mA DC current at the rectifier. I think the 112mA DC saturates the core and makes it hot. So i want to order a new filament transformer with a airgap like a SE output transformer. In this case airgapped for 115mA.
Is this a good idea, or am i wrong ? Thank you


HollowState 27th April 2009 04:52 PM

Hi Johann,

Filament and/or power transformers do not use or require an air gap. If your transformer is running very warm or what seems too hot, it may be a normal temperature rise. It may also be that the unit in question is inexpensively made. Or it may just be under rated. I don't think you can even buy an air-gapped power transformer. Make sure you are using one of the correct rating for your power application as well as your mains frequency.

Ty_Bower 27th April 2009 05:01 PM

If you look at the data sheets for smoothing chokes from Triad Magnetics, they indicate they are "power supply filter chokes having a core with an air gap which prevents saturation at maximum current."

AndrewT 27th April 2009 05:25 PM

are you rectifying and smoothing the transformer output?

The maximum draw from a 2A transformer is 1Adc, if you use a capacitor input filter.
At this value the transformer will run at it's maximum temperature.
It is usually recommended that continuous DC current be kept below 50% of the maximum.
That equates to <250mAdc.

What is your total filament load from this transformer?

tomchr 27th April 2009 05:55 PM

I see you are in Germany, thus, would have 50 Hz mains frequency. Could it be that you are using a cheap transformer made for 60 Hz operation?

The cheap 60 Hz transformers have less iron in them as B field doesn't get as strong as for 50 Hz (16.6 ms period vs 20 ms for 50 Hz). The B field is the integral of the input voltage (i.e. the area under the sine wave curve). Thus, lower frequency --> higher B field = closer to transformer saturation.

You can use a 50 Hz transformer on 60 Hz mains but not the other way around. Many of the higher quality "60 Hz" transformers are really made for 50 Hz operation. That way they can be sold internationally.

As others say - you definitely don't want a power transformer with an air gap.


leadbelly 27th April 2009 06:51 PM

What everybody already said, plus a true NOS labeled GZ32 draws 2.3A, not 2.0A.

EDIT: Oops, that's not quite true, it depends on mfg. I think a Mullard is 2.3A.

Lauscher 28th April 2009 01:54 AM

I try to decribe it once again:
I use a separate EI-core power transformer for the filament of the rectifier. It is not cheap or underrated. It is oversized and from AE Europe in Holland.
I have tested the transformer with a 2.5 Ohm 30Watt wirewound resister. The transformer remained total cool after a hour test.
My Ultron GZ32 has the same cold DCR like my Brimar 5R4GY for filament. Philips datasheet for GZ32: filament 5V - 2A
The power transformer for the rectifier filament has a center tap. That have i used with the direct heated 5R4GY for the B+ voltage. The indirect heated GZ32 cathode is connected to one side of the filament. So for the GZ32 i have used that side of filament for B+ voltage.
The heat problem is the same with both rectifiers.
With the direct heated 5R4GY the filament is the cathode. When i take the 292VDC B+ from the center tap of the rectifier filament transformer, the 115mA current runs through the whole filament winding. I think 115mA current makes a lot stress for the iron core. And this saturation might be the cause for the hot iron core.
As i said i my first post, in use the iron core gets hot, but the coil remains cool.
Most people have the filament winding and high volt winding on the same big transformer. That big core might not saturate so much.
Sorry if i make it too complicate, i try to describe it with my school english.
My idea: the B+ DC voltage and current (from rectifier cathode) that runs through the filament winding, might saturate the core, if there is no airgap. Or am i wrong ? Thank you.


HollowState 28th April 2009 04:06 AM

I understand exactly what you are describing. The transformer stays cool with a resistive load drawing 2 amps. But the core gets hot when used to power the rectifier filament and produce B+ voltage.

If you take the B+ voltage from the end of the winding (pin 5 on the GZ32) and not the CT, there should be no B+ current flowing through the secondary winding. It's potential will be raised to 292 volts, but that will be all that happens. Insulation rating is your only concern. The only closed loop circuit on the secondary is the tube filament.

Now, if the core gets hot when this condition is met, something is wrong with the transformer. (or your wiring) Unwanted DC flowing through the secondary could saturate the core. But in a properly wired and working circuit, that should not be happening. Do you know where the transformer was made? Was it made in China? :dead: It may have an unwanted internal leakage path.

ray_moth 28th April 2009 04:16 AM


Firstly, there is nothing wrong with your English.

Secondly, in the case of the indirectly heated rectifier, the rectified direct current will not run through the filament winding because you are taking it from the correct cathode pin.

Thirdly, in the case of the directly heated rectifier, although there will be DC flowing through the filament winding it will flow in opposing directions, from each end of the winding to the center tap. This will effectively cancel the DC magnetic field.

So, in either case, there will be no net DC flowing through the filament winding. You will not find power transformers with gapped cores, other than the tiny gap incorporated into Lundahl's C-cored power transformers and maybe 1 or 2 other examples, but that's to guard against DC in the incoming mains (to the primary).

For whatever reason, it would appear that the load on the secondary of your filament transformer is causing its core to overheat. This suggests that the transformer is underrated for the job you want it to do. Maybe that should not be the case, especially considering that you tested it with a 2.5 ohm resistor, but it does seem that way.

Lauscher 28th April 2009 10:58 PM


Thank you for the answers.


All times are GMT. The time now is 01:42 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright ©1999-2017 diyAudio