Output transformer question

[Doogie]

Hey,


I'm building my first SE amp from the iron out of an old garbage find console amp (type unknow). The original amp had 9 pin output tubes so I'm assuming they were 6bq5's. Wiring on the empty sockets seem to confirm this. I'm also assuming the output trannys are around 5k primary as the power transformer puts out 184 VAC RMS. I'm using 7c5 output tubes in the new version, because I have a bunch of them.


My question is about the output transformers they measure 1.5” H, 2” W, and the coil is about 1.5” thick (5 watts?). The primaries measure 285 Ohms.
The secondaries I don't understand (newbie here). There are three wires on the secondary, Gray, Green and Black. Measuring with a Fluke DVM I get:
Gray to Green .3 Ohm
Gray to Black .5 Ohm
Green to Black .7 Ohm


On the original chassis Gray and Green wires connected to a speaker socket and the Black wire was soldered directly to the chassis ground. The Black wire does not connect to the frame of the transformer. I haven't seen this before. Anyone have any ideas on why it was wired this way?


Thanks in advance.
Doogie

 

[kevinkr]

Presumably black is the cold end of the secondary and the other two wires are either 4 and 8 ohms or 8 and 16 ohms (first is more likely IMHO based on dcr..)

FWIW I don't get your comment about the black wire.. I've been doing this for way more than 30yrs and have seen very few OPTs with wires soldered to the frame with the exception of some very cheap SE trannies integrated on the basket of an even cheaper speaker.

 

[korey]

I have amps where the black is soldered to the chassis. Its just a ground wire where the chassis is negative anyway.

 

[century tek]

I've seen this too! The black is common ground, so why can't it be the negative ground also. Doesn't seem to have any effect if its grounded to the chassis or not!

 

[artosalo]

I'm also assuming the output trannys are around 5k primary

There is no need to guess what the transformes are. You can simply measure these.

Supply few volts at, say 400 Hz to the primary and take the readings from secondary outputs. (black to grey, black to green and also gray to green)

Then you can simply calculate the turns ratios of all three output combinations.
The reflected anode impedance can then be calculated:
(turns ratio * turns ratio * speaker impedance)

Let's assume that the turns ratio (also voltage ratio) from green to black output is 25. Then the anode impedance is 25 * 25 * 8 ohms = 5k.

I assume that your trafos have 4 ohms and 8 ohms outputs.
Grey to Black is for 4 ohms and Green to Black is for 8 ohms.

 

[Doogie]

Hey,


Wow, thanks much for all the responses. What a wealth of knowledge here! I guess I didn't make myself very clear, or I don't know what I'm doing (more like it), so I drew a crappy schematic for clarification. Given the measurement of the coils in the output transformer this was the way the original transformers were connected. Still seems strange to me that the bottom end of the unused coil connects to ground. Shouldn't the Gray wire connect to the Star Ground , Green wire to + on the speaker and the Black wire left floating?


Sorry to trouble you again.
Doogie

 

[kevinkr]

Hey,


Wow, thanks much for all the responses. What a wealth of knowledge here! I guess I didn't make myself very clear, or I don't know what I'm doing (more like it), so I drew a crappy schematic for clarification. Given the measurement of the coils in the output transformer this was the way the original transformers were connected. Still seems strange to me that the bottom end of the unused coil connects to ground. Shouldn't the Gray wire connect to the Star Ground , Green wire to + on the speaker and the Black wire left floating?


Sorry to trouble you again.
Doogie

Not sure where you got this idea, but likely no.. Measurements with a simple generator will tell you the ratio and from that you can make an educated guess.

The bottom of the coil is ground, most likely there was feedback from one of the taps and this would be referenced to the grounded black tap.. There was probably some connection between the speakers and this point as well.

A schematic of the original amplifier would be most useful.

When you referenced the black wire did you mean to say chassis instead of (transformer) frame in your post on said subject? To me there is a world of difference, perhaps not so much to some of the others here. As an engineer I read things very literally.

 

[Doogie]

Thanks for your patience Kevinkr,
As always when moving into a new culture learning the a new lexicon and semantics confounds the ability to be clear.
My reference to the the Black wire not being connected to the frame of the transformer came from seeing a isolation transformer with a shield wire between the coils that connected to the chassis. I tried to be clear that this was not the case on this transformer.

I don't have a signal generator so I downloaded a sine wave generator program to my laptop. Set it to 400hz and with all the volume levels cranked measured .812 VRMS out of the headphone jack with a Fluke 70 DMM. The results were similar on both the transformers.
Gray - Green = .017 VRMS
Gray - Black = .042 VRMS
Green - Black = .060 VRMS



Hopefully using the formula correctly eg:.
.812 / .042 = 19.3
19.3 * 19.3 * 8 = 2980
We get:
Gray – Green 18248
Gray – Black 2980
Green – Black 1461
Unfortunately there is not schematic for the amp. I pulled the chassis from the garbage and reverse engineered the empty sockets to find the probable tube types. I was pretty careful when I removed the output transformers to take notes. . My notes indicate that the Gray and Green wires went to a speaker socket and the Black wire was soldered to the chassis. The numbers maybe off due to the .812 VRMS input signal.
Yurs Doogie

 

[boywonder]

I'm assuming they were 6bq5's. Wiring on the empty sockets seem to confirm this. I'm also assuming the output trannys are around 5k primary

FWIW, 6BQ5/EL84's typically use 8K/10K output transformers.

 

[kevinkr]

FWIW, 6BQ5/EL84's typically use 8K/10K output transformers.

Believe these are SE transformers in which case 5K would be pretty typical for a 6BQ5..

 

[kevinkr]

Thanks for your patience Kevinkr,

I don't have a signal generator so I downloaded a sine wave generator program to my laptop. Set it to 400hz and with all the volume levels cranked measured .812 VRMS out of the headphone jack with a Fluke 70 DMM. The results were similar on both the transformers.
Gray - Green = .017 VRMS
Gray - Black = .042 VRMS
Green - Black = .060 VRMS



Hopefully using the formula correctly eg:.
.812 / .042 = 19.3
19.3 * 19.3 * 8 = 2980
We get:
Gray – Green 18248
Gray – Black 2980
Green – Black 1461
<snip>
Yurs Doogie

I think you are part of the way there. 400Hz was a reasonable choice IMHO as the primary Z should be sufficient for accurate voltage ratio measurements. I made a couple of assumptions and reworked your calculations.

• I assumed that Black-Grey was the 16 ohm tap
• I assumed that Black-Green was the 8 ohm tap

Taking the 16 ohm case:

812/60=13.53 for the turns ratio, squaring it and multiplying by the suspected tap impedance = 183 * 16 ohms and results in a primary Z of 2928 ohms

Taking the 8 ohm case:

812/42=19.33 for the turns ratio, squaring it and multiplying by the suspect tap impedance = 373.8 * 8 ohms and results in a primary Z of 2992 ohms..

Within experimental accuracy and typical winding tolerances I think the results are consistent, and indicate that what you have is a transformer with a 3K primary impedance.

These on the face of it probably are not the best match for a 6BQ5 in pentode where something on the order of 5K - 7K would seem more reasonable. Note however that with a typical speaker this ought to be OK in triode connection.