50AE--->I found a website explaining briefly tube amplifiers.
The website mentions cross connection of the primary winding sections.
Quote:
Cross connection: The primary windings are alternately cross connected in series around the secondaries (outer P1 - inner P3 - inner P2 - outer P4).
Alternate cross connecting the primary windings enables the physics of the tranny to be closely balanced and the induced magnetic current to be distributed equally into the secondary windings.
The aim is to balance (as close as possible) the primary secondary windings (equally) on both sides of the Center Tap CT. However, balancing can not be perfect, because the inner and outer diameters (circumference) of the windings are different. Wire length and DC resistance of P4 is greater than P1. Increasing the number of primary and secondary interleaving has been the traditional approach to improve the balance and performance of an output tranny.
End of quote.
Now,-the question is: -Does cross connecting apply to single-ended output transformers as well as push-pull?
The website mentions cross connection of the primary winding sections.
Quote:
Cross connection: The primary windings are alternately cross connected in series around the secondaries (outer P1 - inner P3 - inner P2 - outer P4).
Alternate cross connecting the primary windings enables the physics of the tranny to be closely balanced and the induced magnetic current to be distributed equally into the secondary windings.
The aim is to balance (as close as possible) the primary secondary windings (equally) on both sides of the Center Tap CT. However, balancing can not be perfect, because the inner and outer diameters (circumference) of the windings are different. Wire length and DC resistance of P4 is greater than P1. Increasing the number of primary and secondary interleaving has been the traditional approach to improve the balance and performance of an output tranny.
End of quote.
Now,-the question is: -Does cross connecting apply to single-ended output transformers as well as push-pull?
The HP 200CD SG does that to cover 5 Hz to 600 KHz. But not at the same time as required by an amplifier.🙂
Cross coupling is intended for PP class AB or B stages output transformers, where one half primary gets inactive due to one tube shutting off. This can cause high leakage inductance when each 1/2 primary half is physically far away from secondary layers. Worst case scenario happens with double bobbin transformer, Lundahl style, if 1/2 effective primary is located at the first bobbin, and the second primary on the second bobbin. During tube shut off, that would leave a whole secondary package quite far away from the effective primary.
Cross coupling isn't necessary for class A only SE/PP operation.
P.S. In double bobbin power transformers designed for C.T. full wave secondary, high ringing can be observed. From my memories, Thomas Mayer found out high ringing with Lundahl power transformers and I'm positive the reason is the lack of cross coupling. He observed much less ringing by using a bridge rectifier, where all secondaries are active all the time. But he drew the wrong conclusions by blaming the rectifier topology. For this reason, I always split the CT full wave secondary of my double bobbin power transformers into 4/4s and cross couple them. Lower leakage inductance always results in lower winding noise, quieter transformer operation.
Cross coupling isn't necessary for class A only SE/PP operation.
P.S. In double bobbin power transformers designed for C.T. full wave secondary, high ringing can be observed. From my memories, Thomas Mayer found out high ringing with Lundahl power transformers and I'm positive the reason is the lack of cross coupling. He observed much less ringing by using a bridge rectifier, where all secondaries are active all the time. But he drew the wrong conclusions by blaming the rectifier topology. For this reason, I always split the CT full wave secondary of my double bobbin power transformers into 4/4s and cross couple them. Lower leakage inductance always results in lower winding noise, quieter transformer operation.
Last edited:
50AE---> Thank you for the explanation, since I had planned to wind a power transformer with a split bobbin.
Now I know what to do and what not to do.
Now I know what to do and what not to do.
Many people prefer split bobbin power transformers due to the lower primary to secondary capacitance.
The drawbacks however are:
-More prone to winding buzz when feeding into capacitor input filters. Good vacuum impregnation and potting become more mandatory.
-A lot of sag due to high leakage inductance.
-Lower leakage+rectifier capacitance resonance frequency with higher Q (if that's important for some people)
Note that these drawbacks can be also advantages, depending on your application.
Some people stress on the importance of low primary to secondary transformer capacitance for audio equipment. I however, do not understand the need for so much hustle and overthinking on this. I prefer tighter coupled transformers
The drawbacks however are:
-More prone to winding buzz when feeding into capacitor input filters. Good vacuum impregnation and potting become more mandatory.
-A lot of sag due to high leakage inductance.
-Lower leakage+rectifier capacitance resonance frequency with higher Q (if that's important for some people)
Note that these drawbacks can be also advantages, depending on your application.
Some people stress on the importance of low primary to secondary transformer capacitance for audio equipment. I however, do not understand the need for so much hustle and overthinking on this. I prefer tighter coupled transformers
Exactly that is what I found, see:
https://www.diyaudio.com/community/...without-global-nfb.384100/page-5#post-6991884
One rule of thumb I know is.
-Parallel secondary layers that have high capacitance difference between them related to the primary. In a SE transformer, these would be the top and bottom secondary layers.
Example P-S1-P-S2-P-S3-P-S4-P type of interleaving, SE transformer.
Each secondary represents 1R output impedance relative to the primary impedance.
Connect S1 and S4 in parallel for one 1R package.
Connect S2 and S4 in parallel for another 1R package.
Connect both in series for 4R secondary impedance. Ground the first package for best performance.
For 9R, keep the S1 and S4 in parallel
Connect S2 and S3 in series, this is your 4R package.
Then connect the in series to get 9R
Not sure where you got that from. I never used Lundahl power transformers.... Or maybe you mean a different Thomas
Hi Thomas,
Some years ago, I followed your posts quite thoroughly with great enthusiasm. Here are some posts, I believe written by you on the Audio Asylum, dating from 2011, where you mentioned abandoning the FWCT topology due to observations of nasty spikes and less smooth waveform, compared to the Graetz 4 diode bridge
https://www.audioasylum.com/cgi/vt.mpl?f=tubediy&m=194873
Now I might be wrong on my conclusions of which exact power transformer you have tested these findings, as the phenomena can also happen to a slightly lesser extent on a single bobbin transformer, but I'm pretty positive this phenomena happens significantly on a double bobbin power transformer such as the Lundahl or any other double bobbin transformer, where no cross coupling between the windings exist.
I am not intending criticism by any means, as I aim to share my findings in order to reveal more light on unexplained and vague phenomena.
P.S. I'm not quite certain, but I remember observing an article containing some oscilloscope waveforms of FWCT vs bridge and I was associating them coming from your blog.
Some years ago, I followed your posts quite thoroughly with great enthusiasm. Here are some posts, I believe written by you on the Audio Asylum, dating from 2011, where you mentioned abandoning the FWCT topology due to observations of nasty spikes and less smooth waveform, compared to the Graetz 4 diode bridge
https://www.audioasylum.com/cgi/vt.mpl?f=tubediy&m=194873
Now I might be wrong on my conclusions of which exact power transformer you have tested these findings, as the phenomena can also happen to a slightly lesser extent on a single bobbin transformer, but I'm pretty positive this phenomena happens significantly on a double bobbin power transformer such as the Lundahl or any other double bobbin transformer, where no cross coupling between the windings exist.
I am not intending criticism by any means, as I aim to share my findings in order to reveal more light on unexplained and vague phenomena.
P.S. I'm not quite certain, but I remember observing an article containing some oscilloscope waveforms of FWCT vs bridge and I was associating them coming from your blog.
Last edited:
Split-bobbin transformers can be used to good effect, even if it is somewhat irregular:
https://www.diyaudio.com/community/...le-supply-on-a-shoestring.230360/post-3376476
https://www.diyaudio.com/community/...le-supply-on-a-shoestring.230360/post-3376476