1. Win a Broadcaster or one of 3 Teles! The annual Supporting Member Giveaway is on. To enter Click Here. To see all the prizes and full details Click Here. To view the thread about the giveaway Click Here.

Output transformer taps

Discussion in 'Shock Brother's DIY Amps' started by Jewellworks, Sep 4, 2020.

  1. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    3,413
    Joined:
    Dec 30, 2019
    Location:
    california
    As tubegeek said *educated guess*.
    tubegeek may be right that the OT may not be well.

    The 47.3 winding ratio reading you had makes me wonder. (More educated guesses. (Well maybe just guesses, let's leave the educated part out for now. Ha.))

    If
    it is assumed this OT is rated as a 4400 primary. That 47.3 winding ratio puts it in the ballpark with a 2ohm load. 2ohm loads were used back in the day. The Fender Concert and Bassman are two examples. A multi speaker PA may have run with 2ohms.
    Anyway, I wonder if this OT was manufactured as a *universal* trannie. Something that could cover several different applications.
    --------------------------------------------------------

    Assuming a 2 x 6L6 and assuming this Halldorson OT has a 4400 primary... (BTW the 33.7v secondary you measured with 100v on the primary works out to be a 4400 primary with 500ohm secondary.)

    Crunching some numbers with 100VAC on the primaries, voltages on the secondaries would be as follows:
    2.13v for a 2ohm load.
    2.70v for a 3.2ohm load
    3.01v for a 4ohm
    4.26v for a 8ohm
    5.84v for a 15ohm
    6.03v for a 16ohm load

    After checking the OT as tubegeek said, if it seems to be OK, put 100VAC on the primaries. Try to find secondary pairs that give voltages close-ish to the voltages listed above. As I said, one pair you found seems to be in the ballpark for a 2ohm load on a 4400 primary.

    More guesses...
    That amp appears to have some big iron. Big output transformers have the ability to use the next higher or lower speaker impedance without breaking a sweat so even if it only had a 2ohm secondary it may have been used for a 2ohm or 4ohm speaker array.
     
    Last edited: Sep 5, 2020
  2. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    3,413
    Joined:
    Dec 30, 2019
    Location:
    california
    Assuming the OT is functioning OK.
    Hmm. What if the wrong side of the secondary was grounded?

    Try using the 33.70 tap as the reference for all the other taps. What readings do you get?
     
    tubegeek likes this.
  3. telemnemonics

    telemnemonics Telefied Ad Free Member

    Age:
    61
    Posts:
    26,674
    Joined:
    Mar 2, 2010
    Location:
    Maine
    Seems like I've seen pre 1950s PA amps with outputs labeled 4 ohm, 8 ohm, 16 ohm and "70v line"?
    Not sure what for or if that applies to the odd secondary tap, just a thought.
     
  4. Nickfl

    Nickfl Friend of Leo's

    Posts:
    2,554
    Joined:
    May 24, 2016
    Location:
    Florida
    70v is for PA or commercial music systems with long wiring runs. They use a matching transformer on each speaker to match the actual speaker load. Basically it lets you have really long wiring runs without having to use a huge gauge of wire because you have lower current at higher voltage. They are still used today in stores and restaurants where it allows for easy installation of non stereo sound.
     
    telemnemonics and tubegeek like this.
  5. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    168
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    If it was from a PA amp then it probably has terrible low frequency response. I wonder if the readings would make more sense if tested at 1kHz instead of 60Hz?
     
    Asmith likes this.
  6. Jewellworks

    Jewellworks Tele-Meister

    Posts:
    351
    Joined:
    Mar 2, 2020
    Location:
    Orlando
    It's been a busy day...
    On the primary, I get equal resistance from the center tap to both coil leads, and the same when I apply voltage. It's halved at the center tap.
    As for the 100v using different common and taps - anything close? approach suggested by @Lowerleftcoast , YES I came up with some combinations that would give me 2, 3.2, 4 and 8 ohms! So assuming your calculations are correct, I can use this!!?? That would be awesome.
     
    Lowerleftcoast likes this.
  7. Jewellworks

    Jewellworks Tele-Meister

    Posts:
    351
    Joined:
    Mar 2, 2020
    Location:
    Orlando
    @mrriggs -i hear ya about the low freq response. ill have to put in something and find out tho... i can always buy something else if it sounds like a tin can.

    @Lowerleftcoast - there is another listing in the Halldorson catalog that has 2 6L6 PP, 4,8,16 and 500ohm secondaries, but has a primary impedance of 6600. what does that do to your calculations?

    and just what ARE your calculations? what are you doing to come up with these numbers??
     
  8. Jewellworks

    Jewellworks Tele-Meister

    Posts:
    351
    Joined:
    Mar 2, 2020
    Location:
    Orlando
    aaannnddd for what its worth, between 5/3 is 3v= 4 ohms, 5/4 is 4.24v= 8 ohms, 5/6 is 2.12v= 2 ohms, and 2/3 is 2.78v= 3.2 ohms, according to @Lowerleftcoast 's math wizardry (based on a primary impedance of 4400 ohms)
     
    tubegeek likes this.
  9. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,882
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    Hmm. That seems better! So 5 is the common then?
     
  10. Jewellworks

    Jewellworks Tele-Meister

    Posts:
    351
    Joined:
    Mar 2, 2020
    Location:
    Orlando
    Not sure that's exactly true, because I get something useful between 2 and 3. But nothing adds up between 5/2 or 5/1.
    However, there was an odd lead going from tap 5 to what -I can only guess- was a speaker terminal. (A very strange 5 pin connector I've never seen before) And then it jumped to the other one. So tap 5 is "important" for something... But it wasn't grounded
     
    Last edited: Sep 6, 2020
  11. Jewellworks

    Jewellworks Tele-Meister

    Posts:
    351
    Joined:
    Mar 2, 2020
    Location:
    Orlando
    Is there a way to check the primary impedance? I know the impedance reflects the output load, so it changes depending on the load. But if I put a 4ohm load -say a 5 watt, 4ohm resistor- on what appears to be the 4 ohm taps, can I use an ohm meter and read 4400 across the primary? Is it across the main leads? Or the center tap and either lead, that should read 4400? Resistance isn't impedance, so I don't think that'll work... It also needs a working voltage... 350, 400 v...
     
  12. Nickfl

    Nickfl Friend of Leo's

    Posts:
    2,554
    Joined:
    May 24, 2016
    Location:
    Florida
    I guess you haven't found a schematic for the amp this came from? Those don't tend to have transformer impedance values listed, but they can still help a lot when you are trying to map the windings.

    Another thing to consider, I've seen a couple of radio/stereo pull OTs that had the secondaries wired in series like some multi voltage PT primaries are. For example, one amp had blue, yellow, black, and red secondary wire colors. I pulled the transformer without paying that much attention to the wires and only later realized that the black and red wires were tied together on a terminal strip attached to be transformer frame that was not actually itself grounded, so neither of them was a common wire but without the two tied together none of the winding ratios made sense.

    As for your question about testing with a resistor, no I don't think there's any way to do that at least not without some kind of specialized meter like you might use for measuring a choke perhaps.

    Keep in mind that at the end of the day it doesn't really matter what the transformer was built for, it matters what you were going to use it for. So, the winding ratio is all that matters and if you can find a ratio that will give you the primary impedance you want from the speaker you want to use thats what matters. Conversely, sometimes old radio stereos etcetera were meant to operate at lower voltages and with significantly lower primary impedances than you might normally find in a guitar amp and it can be hard to then figure out a way to use it with a commonly available speaker that will give you the primary impedance you really want for your guitar amp project.
     
    Lowerleftcoast likes this.
  13. Jewellworks

    Jewellworks Tele-Meister

    Posts:
    351
    Joined:
    Mar 2, 2020
    Location:
    Orlando
    @Nickfl I drew up a schematic of how this amp was wired before I took it all apart, in case something was wired different that I might need to know about. But otherwise, no schematic. This was a PA amp that someone hacked together many years ago, as the entire top of the chassis was cut way, and a new top was welded into place. It came with old school, big bottle 6L6s (not modern 6L6GCs, but plain old 6L6s) . The power transformer is quite ENORMOUS. Really heavy iron, (weighing in at almost 10 lbs!) and rated at 337.5 - 0 h 337.5 on the 2ndary mains. So it's got high power for days. I didn't see any OT leads tied together. Just tap 1 tied to chassis ground, and tap 5 going to a bizarre speaker connector or whatever it is... I can only assume it ran well for years until it was replaced with something modern and more efficient
     
  14. Nickfl

    Nickfl Friend of Leo's

    Posts:
    2,554
    Joined:
    May 24, 2016
    Location:
    Florida
    Another thing I've seen to consider is that sometimes transformers will have a negative feedback tap of some sort, or even two separate secondary windings, one of which was meant to be used for feedback. If you've got a weird lead that doesn't make sense it could have been something like that, again this is where a schematic really helps but if this was somebody's one-off project obviously all bets are off. Do you get continuity between all of the secondary leads when measuring resistance?
     
    Lowerleftcoast likes this.
  15. Jewellworks

    Jewellworks Tele-Meister

    Posts:
    351
    Joined:
    Mar 2, 2020
    Location:
    Orlando
    @Nickfl , yes I get continuity between all the leads on the 2ndary. All reading in the 9.3 to 9.6 ohm region, when tap 1 is the common. (The lead that went chassis ground). When I use ANY of the other taps as the common, it's .3 or .4 ohms.
    Tap 5, which went to the weird speaker connector, had a 50k NFB resistor going to the cathode of one of the preamp tubes.
     
  16. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    3,413
    Joined:
    Dec 30, 2019
    Location:
    california
    How I went about determining the numbers:
    First, you have given some info about the OT.
    The OT was used in a 6L6 Push/Pull PA amp. I assume it must have worked. What was the speaker load?IDK. Did it have 4, 8, and 16ohm jacks? IDK.
    You said you had some odd winding ratios. I computed the impedance ratio from the winding ratio. Knowing many old OTs with multiple taps include a 500ohm secondary, I checked if a 500ohm secondary tap would reflect a primary for a 6L6 PP amp. I checked the lowest and highest voltages you metered when you put 100v on the primary. The lowest (27.87v) was computed to reflect 6437ohms when the secondary is 500ohms. The highest (33.7v) was computed to reflect 4402ohms when the secondary is 500ohms.
    I assumed the highest voltage of the secondary is the full secondary winding, not a tap. It computed to be 4402:500 reflected impedance ratio, which put this OT in the ballpark to service a pair of 6L6. In addition you found a Halldorson multi-tap OT with a primary of 4400, which made me feel happy.:)
    Likely, I was on to something.:cool:

    So, I now have more confidence Halldorson would assign 4400 as the Primary impedance for this OT. Now the question is... what about the other secondary taps? tubegeek brought up the possibility that this OT may be toast. From what you tested so far, maybe tubegeek was right or maybe not. ??? Let's assume the OT is OK. You did have an initial reading on *a pair* of taps (47.3 winding ratio) that might work after all.

    @tubegeek the following is not a school test that needs grading. Just let Jewellworks think I am a math Wizard. I will be humble. I promise.

    So, given the voltages, you know how to find the winding ratio.
    100v/33.7v=2.967 winding ratio
    2.967 squared = 8.805
    8.805:1 impedance ratio
    You also know how to find the reflected impedance ratio.
    8.805x500=4402.5
    How do I get back to a few volts on the secondaries? I just reverse what I did to get to the impedance ratio.
    Now I have the 4400:500 reflected impedance ratio. I want to find a 4400:8 reflected impedance ratio.
    (x) x 8 =4400
    (x)= 4400/8 = 550
    550:1 impedance ratio
    square root of 550 = 23.45 winding ratio
    100/(y) = 23.45
    100 = 23.45 x (y)
    100/23.45 = (y) = 4.26v
    When 100 volts are applied to the primary, a secondary with 4.26v will show a reflected impedance ratio of 4400:8

    I repeated this for the other reflected impedance ratios.
    Wizardry:rolleyes:
    It doesn't do anything to these calculations.
    With all the taps on this OT, and/or getting a little creative as Nickfl noted, you may be able to arrive at a ~6600 reflected primary. I feel pretty confident this OT was designed to have a reflected primary impedance of 4400.
     
    Jewellworks likes this.
  17. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,882
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    @Lowerleftcoast

    What about the issue another perceptive commenter @mrriggs mentioned though?

    If the 60hz performance of this OT is not an accurate reflection of its actual winding ratio, we may come to another problem.

    The output voltage at 60hz maybe is too small compared to, say, 500hz. So the measured ratio will seem steeper than the "real" ratio. Meaning the REAL ratio will be closer, meaning the primary impedance will be lower than the exciting possibility of 4400.

    I'm not sure about the effect that not loading the secondary while measuring will have though?

    The answer to all of this is of course the same: light the sucker up and see what you get.
     
  18. elpico

    elpico Tele-Afflicted

    Posts:
    1,190
    Joined:
    Sep 14, 2011
    Location:
    Vancouver BC
    Hold up, this is telling you something important. The tap that is 9ohms to the others is absolutely NOT going to be involved in driving your speaker. You'll need a pair of the leads that are under an ohm. It's the voltage between those that you're interested in.
     
    Lowerleftcoast and tubegeek like this.
  19. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,882
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY

    Light bulb moment. Fact. That 9ohm tap1 MUST be for something else like a 500 ohm out, or nfb, it can't be a speaker secondary or you'd lose almost all your output in the DCR.

    Pin 5 is your nfb connection, it had the nfb resistor, so it's a winding dedicated to nfb and it has nothing to do with the speaker secondaries. I was wrong about the doorstop it appears, sorry!
     
    Last edited: Sep 6, 2020
    Lowerleftcoast likes this.
  20. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,882
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    I just did a double-check based on this list:

    100V in, gets these outputs:

    33.70
    31.58
    30.65
    29.45

    33.70 - 31.58 = 2.12
    33.70 - 30.65 = 3.05
    33.70 - 29.45 = 4.25
    33.70 - 27.87 = 5.83

    The ratios between those voltages are almost exactly square root of 2 (1.41) each: 1.43, 1.39, 1.37.

    So the 33.70 measurement was indeed the common terminal. The 500 ohm output sends a big neg FB signal into the cathode, the rest are for speakers.

    I hate when I get my head wedged up in there, it's so hard to get out sometimes.

    @mrriggs - I don't think you have to worry about the 60 Hz bandwidth of this transformer, it seems to be passing that frequency OK - I **think** to confirm this further, we'd need to load the secondary and re-test but so far nothing looks problematic, assuming we want to call this a 4400 ohm primary transformer.
     
    Lowerleftcoast likes this.
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.