reflected load

[jcm-jmp]

hey guys,

its been a while, hope all is well.

ok so i have an old ot i pulled out of an old bogen chb100. It has 4-8-16 ohm taps along with a 32ohm or 25v tap.

I have been trying to figure out what the reflected load is on the primary using each tap.

i hooked 100v to the primary and read 34v, 24v, 17v and 12v

so correect me if im wrong please  

the ot wants to reflect a 4.6k reflected load. another words if you use a 4ohm load on the 4ohm tap it reflects 4.6k to the primary and 8ohm load on 8ohm tap it reflects 4.6k to primary ect.....

it reflects 4.6k to primary allways as long as the proper load is on the proper tap.

now

if my memory serves me, the bogen used 4-7868's in push pull. the tube data says the 7868 want a reflected resistance of 6.6k . How does this relate to the 4.6k reflected load from the properly loaded ot the bogen came with?

it seems to me that 4.6k load is smaller than what the 7868's like to see (6.6k). why would they design an amp with too low of reflected load for the type of tubes being used. or have I totaly fowled up the reflected load theory?

 

[tubeswell]

i hooked 100v to the primary and read 34v, 24v, 17v and 12v

Your readings are puzzling. If an OT reflects 4k6 with a 4R speaker, it would have an impedance ratio of 1150:1, and therefore a VAC ratio of 33.9:1, which means that if you hooked up 100VAC to the primary you'd get 2.95VAC on the 4R secondary tap.

Did you use AC voltage to do the test? (Also, its far more accurate to supply a small (like 5 or 10) VAC to the secondary, and measure the resulting VAC on the primary)

[PRR]

100V in 12V out at 4 ohms does not make sense.

> the tube data says the 7868 want a reflected resistance of 6.6k

That's a suggestion. Not a requirement. We see 2*6L6 loaded with 3,400 to 10,000.

And of course, _if_ 6.6K were good for two tubes, then four tubes should be 3.3K.

Bogen knew their stuff. If you use that OT with the original PT (and rectifier scheme), with four suitably large tubes, it will do 100 Watts again.

[jcm-jmp]

ok i applied 100v on primary via a variac.
the secodary readings were 3v, 4.3v, 5.9v, and 8.4v (sorry for the mistake above)
the turns ratio came up to  33.3:1, 23.25:1, 16.95:1, 11.9:1
which gave me a more or less constant 4k6 reflected z on the primary using the proper load figure on the proper tap.

bogen used 7868's which want 6k6 reflected to 2 tubes fixed bias ab1 push pull. 4 tubes wanting 3k3 reflected fixed bias ab1 push pull.  

so I guess 4k6 is not a big enough difference from 3k3 to fuss about.

yes PRR I agree. I do have the PT as well and I'm planning to build a 100w geeter amp. I just wanted to use el34's rather than 6l6 or 7868 or 7591.

2- EL34's want 3k5 so 4 - EL34's want 1k5 reflected z?  

Also I am try to get a better understanding of reflected z and the effects of the ot design on frequency responce.

[tubeswell]

2- EL34's want 3k5 so 4 - EL34's want 1k5 reflected z?  

Yep - more like 1k7 for 4 if you have 3k5 for 2.  These are nominal numbers. But the principle is the same. Double the current, halve the resistance, for the same amount of voltage.

[PRR]

> secodary readings were 3v, 4.3v, 5.9v, and 8.4v (sorry for the mistake above)

mmmmmmph....

Well, _now_ we have 4.4K-4.6K.

> 7868's which want 6k6

The tubes don't "want" any thing.

The limits are:

The tube's saturation resistance is about 200 ohms, so we want much-more than 200 ohm loading.

We could use 10K or 20K loading (much-much-more than 200 ohms!) _but_ we would not get much power unless the supply voltage were very high, and the data-sheet says 550V max.

Datasheet writers usually can't get too clever. They can't call for absurdly high supply voltages or strange load impedances, or amp-builders will move on to a "more convenient" tube. In fact at the time the majority of hi-fi amps in the 40+ watt range were using a 6.6K part from various OT winders. And 450V is about the limit for common 450v/500V electrolytic caps.

Also a tube-company may want to market some tubes for mass-market low-price and save other tubes for special customers who will pay more without much complaint.

Also note that this data-sheet does NOT show any 50 Watts per pair (100W/quad) condition. And yet the Bogen 100 will do an honest 100W (some industrial customers actually test what they buy). So the Bogen boys went beyond the suggested starting conditions.

BTW: the suggested 44W condition leads to only 11 watts dissipation per tube at full power. It's running 82% of max voltage and 58% of max dissipation... the tube is not strained.

Since 4600/quad is 9200 per pair, yet power is higher, I suspect Bogen raised the B+ over 525V (60W out), then raised load impedance (reduced power) to hit their 50W/pair power goal.

> just wanted to use el34's

So use EL34. You will need a lot more G1 bias than 7868, be prepared for 40V-50V of bias.

> effects of the ot design on frequency responce

Different question. And moot if you use this OT with designed impedances and pentodes. You get essentially all the Hz Bogen designed for, which is ample for nearly any audio purpose.

[jcm-jmp]

AH I get it.

The limits are the power rail voltage B+ in conjunction with the tubes design limits and the loading of the particular tube choice.

a little time with ohms law and the data sheets will get me close enough for geeter amps.

The 200 ohm saturation point is still a puzzle for me, but im sure i will get the math soon enough

but the refelected Z is a very wide range depending on how a designer wants to establish and regulate the power output stage in a particular amp.
now I suspect that this ballancing act between voltage and Z (as well as ot physical design) are going to establish the amount of negative feed back and harmonic distortion levels ?

there again this is a geeter amp not a hifi amplifier  

PRR thank you for you input. Again, you have shed enough light for me to grasp the concept.  

http://www.electro-tech-online.com/custompdfs/2011/12/Audio20Transformers20Chapter.pdf
Im reading this
 

[PRR]

> The 200 ohm saturation point is still a puzzle for me

[PRR]

> Im reading this

Bill knows WAY too much about transformers.

I suggest you start with Radiotron Designers Manual, first the 3rd and then skim the 4th.