Testing output transformer

Discussion in 'Tube Audio' started by Ziradog, Dec 7, 2018.

  1. Ziradog

    Ziradog AK Subscriber Subscriber

    Messages:
    1,065
    Location:
    Apex, North Carolina
    I'm sure there is a decent thread on how to test an output transformer that is loose (not in an amp), but I have tried several searches without success. Can someone point me in the right direction? Preferably one that does not require specialized equipment. Thanks.
     

     

    Please register to disable this ad.

  2. Tom Bavis

    Tom Bavis Audiophool Subscriber

    Messages:
    9,089
    Location:
    Macedon NY
    Ohmmeter is a start. Primary will measure maybe 5% of its AC impedance (so a 10K transformer may be 500 Ohms DC resistance) Secondary will probably read zero, as most meters can't resolve tenths of an ohm. If you don't know the impedance, here's how to measure: https://www.radioremembered.org/outimp.htm
     
    Ziradog likes this.
  3. BinaryMike

    BinaryMike Pelagic EE Subscriber

    Ziradog likes this.
  4. trobbins

    trobbins Well-Known Member

    Messages:
    521
    Location:
    Melb, Australia
    Ziradog, what testing did you want to do? What test equipment do you have? Were you aiming to determine what an OT is good for - eg. frequency response, power rating, distortion - or just impedance ratio? Do you know it is an output transformer, and if so then what do you know about it?
     
    Ziradog likes this.
  5. Ziradog

    Ziradog AK Subscriber Subscriber

    Messages:
    1,065
    Location:
    Apex, North Carolina
    Trobbins, I have found a transformer that came from a Pilot 240 to replace the generic OPT that is in one channel my 240 now. It looks rough, and I want to make sure it is functioning more or less properly. So I know it is an OPT and what the hook ups are supposed to be. I know the impedance ratio. Tools, I have a DMM & a simple oscilloscope, variac, not much else.
     
  6. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    37,825
    Location:
    Southern NJ
    voltmeter and a variac will get it done well enough. Basically if its not shorted or open, and the impedance rings out right, its most likely fine.
     

     

    Please register to disable this ad.

  7. Ziradog

    Ziradog AK Subscriber Subscriber

    Messages:
    1,065
    Location:
    Apex, North Carolina
    Sorry for the delayed follow up, I got sidetracked on other projects. For the method in BinaryMike's post, could you use a Variac instead of a filament transformer to drive the secondary windings? The transformers are out of the donor chassis.
     
  8. trobbins

    trobbins Well-Known Member

    Messages:
    521
    Location:
    Melb, Australia
    Much safer to use an isolating transformer (ie. filament transformer or similar) than a variac.
    You can try and measure the primary winding inductance, eg. with 5 to 20Vac applied across half the P-P primary, and your meter measuring mA AC. An AVO was a common meter to use back in the 40-50's and it had just enough deflection on most sensitive current scale to get a measurement on h-fi OPT's that have large inductance. Rather than try and get an absolute measurement, the better aim would be to compare the unknown OPT with a generic OPT to see if they were similar. Primary inductance is an indicator of low frequency performance and whether you may have instability issues at the low end.
    You can try and measure frequency response if you can rig up a good soundcard and amplifier with known response out past 20kHz.
    A turns-impedance calculator spreadsheet with measurement notes that may help: http://dalmura.com.au/projects/OT calcs.xls
     
    Ziradog likes this.
  9. BinaryMike

    BinaryMike Pelagic EE Subscriber

    I've probably done it once or twice with a barefoot variac, but as trobbins said, it's much safer to use an isolating transformer. If you only have a higher-voltage transformer, e.g. 12 V or 24V, then you could use it with a variac to get something in the 5~7V range for my OPT test method.
     
    Last edited: Jan 7, 2019
    Ziradog likes this.
  10. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    37,825
    Location:
    Southern NJ
    I use my variac plus the iso transformer. I adjust it until I get 1 vac on the secondary and measure the primary input voltage. The math follows from there. I actually knocked together a simple spreadsheet to do that part for me but its nothing a calculator won't tell you.
     
    Ziradog likes this.
  11. trobbins

    trobbins Well-Known Member

    Messages:
    521
    Location:
    Melb, Australia
    Strangely, I extend the filament voltage from an old valve tester for this application more than for actual valve testing - most testers provide a wide range of easy to change heater voltages, and it has a hefty current rating and is isolated.

    If you are testing a PP transformer, then I recommend applying the test voltage to one half of the primary winding (CT to anode), and take a measurement from the other half of the primary, and the secondaries - that is a little more accurate. You also need to be mindful of how accurate your meter is at the relatively low AC voltages that occur on secondaries (eg. a reading of 0.9V is effectively +/- 10% plus the meter error).
     
    Ziradog likes this.

     

    Please register to disable this ad.

  12. Ziradog

    Ziradog AK Subscriber Subscriber

    Messages:
    1,065
    Location:
    Apex, North Carolina
    OK then. Maybe the power transformer from the donor unit will be good & I can use that.

    trobbins, it is a push pull transformer. The schematic says the resistance (not impedance) on the primary should read 250 ohms on each side, I got 228 and 208. I'm not happy with that difference, so I may pull the other (uglier) OPT from the donor and see what that measures. Should that concern me?
     
  13. primosounds

    primosounds Powered with pure tube sounds. Subscriber

    Messages:
    3,018
    Location:
    Terra, 21st century CE
    I buy a lot of trannies so i am always testing them. What i have noticed about PP, is that the DCR between the 2 halves are rarely close. The difference might be worse on the "economy" ones, but, your measurements do not indicate a defect. Also if the 2 different trans did not come off the same assembly line then it may account for why your readings won't be close to 250 ohms.
     
    Ziradog likes this.
  14. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    37,825
    Location:
    Southern NJ
    difference in ohms end to middle is not too unusual. The inner half of the winding requires less feet of wire. You're in the ballpark of the spec. Mostly its there to rule out obviously open or shorted transformers.
     
    Bill Ferris and Ziradog like this.
  15. trobbins

    trobbins Well-Known Member

    Messages:
    521
    Location:
    Melb, Australia
    Only the most fastidious of PP OPT's typically have equal primary half winding resistances. That level of OPT concern and construction started with the Williamson amplifier, where primary DCR is typically within 1 ohm for each half.

    With respect to what that resistance affects, it adds in to the plate resistance of the output stage tube and that summed resistance interplays with the low and high corner frequencies of the output stage. You didn't provide a link to a 240 schematic - it seems to be a pentode output stage (not a triode format) - so any DCR difference in the OPT windings is absolutely swamped by the pentode plate resistance, and the relatively huge differences that would typically exist between tube plate resistance of the same make/model tube. So there is no credible reason to be concerned.
     
    Bill Ferris and Ziradog like this.
  16. PakProtector

    PakProtector AK member

    Messages:
    3,731
    Location:
    Ann Arbor, MI
    The power can be tested with a signal generator, a SS amp, an o-scope and some resistors. Drive the secondary, with its rated load in series with the SS amp. Apply a load equal to what is reflected from the secondary. This will take some power dissipation, so proceed with caution as things can be hot, and the AC voltage can get very high. Voltage divide the primary's voltage so it does not crisp the measurement equipment. Apply signal, measure amp output voltage, and the fraction of voltage across the voltage divider transferred to the primary. Plot them.
    cheers,
    Douglas
     
    Bill Ferris, Retrovert and Ziradog like this.

     

    Please register to disable this ad.

  17. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    37,825
    Location:
    Southern NJ
    Pilot 240 is an EL84 pentode amp.
     
  18. Leland47

    Leland47 AK Subscriber Subscriber

    Messages:
    2,717
    There are numerous YouTube vids about it.

    Here is one.

     
    Ziradog likes this.

Share This Page