'65 Deluxe Reverb OT impedance connundrum

[octal]

Help me figure this one out! It seems that all replacement Deluxe Reverb output transformers claim to be 6600 ohms:8 ohms. I was looking at the '65 DRRI schematic from Fender and it shows 174 Vac on the plates of the output tubes, and 11 VAC at the speaker output. By my reckoning, you should double the AC voltage on the plates, so 348V. Divide 348 by 11 to get N, the turns ratio... 31.6V. Square that, it's about 1000, multiply by the speaker impedance (8 ohms) and the reflected impedance comes out to 8K.

So, longwinded explanation aside, what is the proper load impedance for a Deluxe Reverb? 6.6K or 8K?

Nathan

[Dave H]

I've measured the impedance ratio of OTs in circuit like that but with no load and it gives the correct ratio. Perhaps with a load connected the secondary voltage is reduced a bit because of circuit losses giving a calculated ratio of 8k when unloaded it would be 6k6?

Dave H.

[Five_E35]

I was temped to try one in my 5E3 build and found a lot of conflicting information on the net about the DR RI transformer. Many sources list it as 6K6, others as 8K5. I know resistance doesn't matter, but rather than shunt, I bias by dividing the voltage drop by the OT resistance and some searching on the net turned up ~215 ohms per side compared to ~145 ohms for the Weber 6K6. Maybe what you've confirmed is some people actually measured it and others assumed it was the same. I also found conflicting information on whether the original BF & SF versions were the same. Or maybe Fender was trying to calm things down in the RI for the older Russian 6V6GTs?

[MWJB]

7K for a typical DR style OT, maybe plus/minus a few hundred ohms depending on mfr.

Five E35 - Difference in DCR accross the primaries could be more indicative of guage of wire, rather than turns ratio, so it's not a deciding factor.

Octal - Your formula seems flawed, checking the other Fender schems by the same method, reveals that the resulting primary Z uniformly comes out on the high side? E.g. 10K for the 57 Deluxe, which I thought used MM trannies? nearly 5K for the SR, 4.4K for the 57 Twin?

Unplug from the wall, pull the tubes & the speaker, apply 0.5VAC (measured) to the secondary, what do you get at the primary?

[octal]

I'm liking Dave H's theory. The transformer will have losses due to the resistance of the windings, core losses, etc. so the secondary voltage will be less than the theoretical number which assumes an ideal transformer. The loaded secondary voltage will be lower than the ideal lossless value, which will make the transformer appear to have a higher primary impedance.

I kind of wonder if the "true" loaded voltage ratio might be a better indication of what kind of reflected load the tubes will really be seeing than the unloaded value?


Nathan

[MWJB]

Transformers aren't interested in DC resistance, they see AC impedance. You could have 2 OTs with the same turns ratio, one with 30ohms DCR from primary to CT, the other with 150ohms - they will still have the same AC impedances.

Do the test I suggested, what do you get?

[octal]

Transformers aren't interested in DC resistance, they see AC impedance. You could have 2 OTs with the same turns ratio, one with 30ohms DCR from primary to CT, the other with 150ohms - they will still have the same AC impedances.

Do the test I suggested, what do you get?

I think you may have misunderstood me. This is a theoretical question...I am not trying to measure the impedance of an actual transformer I have on hand to test. I know how to do that! I was trying to determine the true specs of the transformer used in the '65 Deluxe Reverb reissue based on AC voltages shown on the schematic.

You are correct, the resistance of the windings has no bearing on the unloaded voltages you'll measure when trying to determine impedance of the transformer. However, the resistance of the windings should effect the voltage the transformer will deliver into a load. A transformer has losses due to the resistance of the windings, the imperfect coupling of the magnetic field into the core, etc....i.e. it is never going to be 100% efficient. Those losses are really only going to show up under load.

Cheers,

Nathan

[MWJB]

Might there not be a simpler answer? The test signal in all the cases quoted is 1kHz, the speaker's actual impedance will be somewhat over the nominal rating at this frequency in many cases, hence a reflected higher primary Z?

[octal]

MWJB,

The schematic lists the test conditions- for power output, it's specced into an 8ohm resistive load, but for the operational AC voltage measurements, it's unclear whether it's being measured into the speaker or a dummy load. I would almost bet that the impedance of the speaker is greater than 8 ohms at 1kHz...it tends to rise due to voice coil inductance, so you might be on to something here.

I think we can both agree that for whatever reason(s) "my" method of calculating transformer impedance ratios is probably inaccurate when the output of the amp is loaded. I'll try some experiments when I have a moment and an amp sitting in front of me.

Nathan