Unequal resistance on output transformer primary

[bobloblaws]

Measure your speaker.Your going to be looking for an OT that has anywhere from 6 to 8k on one side and the other side depends on your speaker if it's 4, 8, or 16 ohms. The 6 to 8k depends on the value of your cathode resistor. http://www.r-type.org/pdfs/el84.pdf

nosaj

The amp was given to me for service without a speaker. The amp has been returned to the owner for evaluation and he asked me about the proper speaker load so it seems he does not have the original speaker.

[bobloblaws]

Sorry, I may have inadvertently deleted something. You asked about checking inside the transformer. Please see my response to Tom.

[nosaj]

The amp was given to me for service without a speaker. The amp has been returned to the owner for evaluation and he asked me about the proper speaker load so it seems he does not have the original speaker.

Then in that case i would choose the classic tone OT I linked. That way whether he gets a 4 or 8ohm speaker he's covered.

I look at the OT like a seesaw. per the datasheet the 2 el84s prove 6 to 8k on one side of transformer and the other side will be wound for whichever speaker is used 2,4,8,16,32 etc.

now using the incorrect speaker load could be akin to either redlining a Yugohttp://www.librarypoint.org/yugo_vuic car all the time or Having a Corvette with a 4cyl engine. One is overworked(meaning stuff dies faster ) and the other is underpowered . So having a good balance allows maximum transforming of signal.
nosaj

[bobloblaws]

Your going to be looking for an OT that has anywhere from 6 to 8k on one side and the other side depends on your speaker if it's 4, 8, or 16 ohms. The 6 to 8k depends on the value of your cathode resistor. http://www.r-type.org/pdfs/el84.pdf
nosaj

So I did some reading on how to determine reflected load etc. (Output Transformer Impedance)and I have a pretty good handle on it. But I still have a few questions. When you talk about 6 to 8K on the primary side, is this based on the Ra-a -> 8.0kOhm specification on the EL84 datasheet you linked to? What exactly does Ra-a stand for, R for resistance obviously, but a-a? The cathode resistor in this case is 150 ohm. How does that affect whether the impedance is closer to 6 or 8K? I also don't understand why we are talking about values in the thousands of ohms (e.g. 8kOhms) but the resistance readings I generally get when measuring across the outer legs of a OT primary are in hundreds of ohms, as in between 100 and 400 ohms.

[g1]

Impedance is to AC, resistance is not the same. DC resistance values are often very different from the impedance a winding of wire presents to AC.

[nosaj]

So I did some reading on how to determine reflected load etc. (Output Transformer Impedance)and I have a pretty good handle on it. But I still have a few questions. When you talk about 6 to 8K on the primary side, is this based on the Ra-a -> 8.0kOhm specification on the EL84 datasheet you linked to? What exactly does Ra-a stand for, R for resistance obviously, but a-a? The cathode resistor in this case is 150 ohm. How does that affect whether the impedance is closer to 6 or 8K? I also don't understand why we are talking about values in the thousands of ohms (e.g. 8kOhms) but the resistance readings I generally get when measuring across the outer legs of a OT primary are in hundreds of ohms, as in between 100 and 400 ohms.

Ra REsistor Anode(Plate) Rk Resistor Cathode(Cathode).
R-a -a I believe is Resistor anode to anode

If you go back to the seesaw theory The seesaw being the transformer. If there is no one on either side You can't know what either side will take without data. But if you know the weight of one person and where the fulcrum is at on the Saw you can figure what the weight of the second person needs to be.

If that analogy doesn't help I'm kinda of stuck. As much as I don't understand Math i am starting to come to an understanding albeit a little at a time.
Hope it helps
nosaj

[bobloblaws]

Ra REsistor Anode(Plate) Rk Resistor Cathode(Cathode).
R-a -a I believe is Resistor anode to anode

If you go back to the seesaw theory The seesaw being the transformer. If there is no one on either side You can't know what either side will take without data. But if you know the weight of one person and where the fulcrum is at on the Saw you can figure what the weight of the second person needs to be.

If that analogy doesn't help I'm kinda of stuck. As much as I don't understand Math i am starting to come to an understanding albeit a little at a time.
Hope it helps
nosaj

"A" for anode, of course, it came to me soon after I clicked "reply". Actually I think I understand now after thinking it through, it would be impossible to directly measure a reflected impedance as there is no actual physical connection between the primary and secondary. Still wondering about the cathode resistor bit though.

[murrayatuptown]

IF one has the ability to measure inductance, something I check as 'one more thing' as a troubleshooting observation, is to measure each winding inductance (low ac voltage test instrument), and short one of the other windings. This is one method of measuring leakage inductance. With a transformed winding shorted, the 'good' measured inductance will drop to almost nothing (depends on the nature of the windings), say from Henries to milli-Henries. Conversely, an abnormally low inductance measurement (helps to have some idea APPROXIMATELY what to expect) with no windings (externally) shorted is a likely indicator a winding short exists...it creates a condition similar to the leakage inductance measurement test.

One caveat is that a piece of test equipment that measures inductance (be it an RLC bridge or digital device that likely is also a bridge but you don't know what's inside) does it at a single frequency and a rather small voltage. You take such measurements on iron-core magnetics, whose normal operating condition is very different, with a grain of salt, in the context of what it can tell you. You can estimate turns ratio to identify which winding is which (if not marked) from the square root of the impedance ratio, which is proportional to the inductance ratio, determine some good/bad answers, etc. But you cannot assume the inductance you measured at 100 mV, 1 V, 100 Hz, 1 kHz or whatever, have anything to do with primary inductance with full winding voltage, and sometimes DC bias, that you care about to figure out low end frequency response.

It's tricky...you sometimes get unexpected measurements...because your method isn't correct for the circumstances.