Reamp: what cheap but good transformer?

[PRR]

Hard telling not knowing, ....

Hey! Have you been reading "my" newspaper?
"Hard Telling Not Knowing each week tries to answer your burning questions about why things are the way they are in Maine — Bangor Daily News,  https://www.bangordailynews.com

the recycle pile where old stereo amps / receivers end up.
Or the "junk" store where they sell old audio equipment for next to nothing.

Audio transformers are so troublesome (response, distortion, cost, weight) that they are almost unknown in consumer gear. Yes, Rob (and I) remember audio transformers in pocket radios and in 12V PA amps, but most of these were not "repeat coils" or had whack impedance or lousy response. And I am in favor of stripping unloved GoodWill gear, but not with much hope.

[taudio]

Thanks for the clarification Rob -- much appreciated!

For the frequency response test, I drove the reamp box with a signal generator that has a 600 ohm output impedance.

[Elijah-Baley]

Back to the Triad TY-250P, I found this:
https://www.talkbass.com/threads/an-inexpensive-audio-transformer-with-diy-projects.1113802/#post-16795141

Does it means that there's the risk I get distortion in the low end by guitar bass, or low tuning hot output guitars?

[kenjib]

Are you ever going to run 20V peak-to-peak through it?

-k

[Elijah-Baley]

Oh, well... 20v seems really too much!
If I don't miss some step: guitar or bass, sound card... Maybe there's no risk to get a distorted signal.
But I need a confirm, please.

[Rob Strand]

Oh, well... 20v seems really too much!
If I don't miss some step: guitar or bass, sound card... Maybe there's no risk to get a distorted signal.
But I need a confirm, please.

The Triad spec is 4.2Vrms (12Vp-p) at 20Hz.

I can confirm after applying some reverse engineering to the transformer that 4.2Vrms is pushing the transformer a little hard but certainly not over doing it.  Perhaps 3.4V rms (10Vpp) would be a good upper limit.  Whereas a more desirable upper drive level to keep the the distortion low would be about 1V rms to 2Vrms (2.8Vpp to 5.7Vpp).

The talkbass plots very much support those conclusions. You can see at 10Vp-p there is an effect at at 5Vp-p nothing is visible.  Be aware, when a frequency plot is affect by distortion like that the distortion is usually noticeable.
https://www.talkbass.com/threads/an-inexpensive-audio-transformer-with-diy-projects.1113802/#post-16795141

The thing is when you back off from the saturation region the distortion also drops quite significantly.

You can see that at 1V rms (0dBV)  the THD is pretty good at Hz, in fact the distortion at 100Hz seems more.
https://www.audiosciencereview.com/forum/index.php?threads/e14-1-1-600-600-cheap-transformers-some-measurements.10438/

You should be able to confidently get down 1Vrms to 2Vrms (2.8Vpp to 5.7Vpp) at 20Hz.

The other thing to keep in mind that with many signals the amount of signal at 20Hz is quite low.  The signal levels at 100Hz to 200Hz will be much higher.   At 100Hz to 200Hz the you can drive the transformer at 5 to 10 times higher levels than at 20Hz.

You can do a test which deliberately drives the transformer hard until you can hear the transformer affect the signal then note that level.  That will give you an idea how much headroom you have in a practical set-up.   You will probably find that it's a lot better than you expected.   You might even find at that point it's the solid state circuits clipping and not the transformer saturating (confirm by listening to the signal going to the transformer).

[PRR]

the distortion at 100Hz seems more.

To first-order, level may go up with frequency. 1V@20Hz, 2V@40Hz, 4V@80Hz, 8V@160Hz. That's why dinky iron is acceptable for communications speech, >300Hz. (Also of course why 25kHz switchers can be so much smaller than 50/60Hz hummers.)

While guitar goes to 82Hz and less, your screaming lead may have little below 300Hz. (But for general re-recording you have to allow for 50Hz.)

[Rob Strand]

the distortion at 100Hz seems more.

To first-order, level may go up with frequency. 1V@20Hz, 2V@40Hz, 4V@80Hz, 8V@160Hz. That's why dinky iron is acceptable for communications speech, >300Hz. (Also of course why 25kHz switchers can be so much smaller than 50/60Hz hummers.)

While guitar goes to 82Hz and less, your screaming lead may have little below 300Hz. (But for general re-recording you have to allow for 50Hz.)

There's a good chance that 120Hz peak on the distortion plot (from the www . audiosciencereview . com link) is 120Hz.  There's a bit of a bump at 60Hz as well.

[Rob Strand]

This document has quite a lot of info,
https://www.jensen-transformers.com/wp-content/uploads/2014/09/Audio-Transformers-Chapter.pdf

Go to pages 10 and 11,
- Figure 17
  Different material behave differently.
  Some increase distortion with level some decrease distortion with level.
  Once you approach the saturation point the distortion increase with level.

- Figure 18
  Distortion decreases with frequency (for voltage drive)

- Figure 19
  Distortion increase with source resistance.
  A higher source resistance allows the the non-linear core inductance
  to affect the voltage at the transformer input terminals.

All this info has to be measured for a given core/transformer.

[Elijah-Baley]

Ok, I'm not so knowledgeable on this subject, I'm learning reading all of you.
Probably I can place an order today or very soon.
Do you think that a TY-250P is good enough for the purpose of my friend?

[Rob Strand]

Ok, I'm not so knowledgeable on this subject, I'm learning reading all of you.
Probably I can place an order today or very soon.
Do you think that a TY-250P is good enough for the purpose of my friend?

It's very hard to do a paper exercise with these small transformers and be 100% certain it will work.  At some point you need to test it.   However, we aren't going in cold.  People have made measurements.   All the evidence looks like it should do the job.

If the output comes from a PC or a USB DAC then it's likely the output will be limited to 2V peak (4Vpp) anyway.

As a bit of icing on the cake you could add a Zobel network across the transformer secondary,

https://sound-au.com/articles/audio-xfmrs.htm

The values of the resistor and cap depend on the transformer.   There's no easy way to choose the value.

RG has 1n + 10k on a Xicon transformer,
http://www.geofex.com/FX_images/TransformerSplitter.pdf

Somewhere along the way  I saw something like 1n + 1k5 on a TY-250P.

[Elijah-Baley]

Maybe I will make an attempt with the TY-250P.
Meanwhile, I was searching some alternative project to re-use it in case I fail with the reamp.
I could open another thread about it.

We still have to see the schematic!
Probably something simple, without a buffer, but I still considering if it really need to included it.

[Rob Strand]

Maybe I will make an attempt with the TY-250P.
Meanwhile, I was searching some alternative project to re-use it in case I fail with the reamp.
I could open another thread about it.

We still have to see the schematic!
Probably something simple, without a buffer, but I still considering if it really need to included it.

I wouldn't think too far ahead until you try it.  There's more positives than negatives for the TY-250P at this point.

The thing is Triad clearly has made some effort to get good a LF response.   To do any better and/or get more drive you will need a larger core.  A 25mm wide core might help but cheaper models could still end-up with a poorer LF response than the TY-250P.  Going up to 35mm to 42mm cores is probably where good LF and drive happens by nature but that's a significantly larger core.   If you ignore price and look at more professional audio transformers you will find they are in the 35mm to 50mm wide region.  The core size isn't the only difference, expensive transformers can also have different winding patterns.

[PRR]

without a buffer,

Then use a Western Electric Repeating Coil. These are older than amplifiers and are over-over-engineered to minimize transformer faults in non-amplified applications.

Yes a power supply and an opamp is a lot cheaper than a repeat coil.

[Elijah-Baley]

If you ignore price and look at more professional audio transformers you will find they are in the 35mm to 50mm wide region.  The core size isn't the only difference, expensive transformers can also have different winding patterns.

Unfortunately, my friend don't want to spend a lot for something he will use not so often. This is the thing. Some reamp on the market are a bit overpriced (of course, it depends case by case). I'd buy Lehle to build something good, or even very good. But this is not the chance.

So, or we can settle we this small transformer or it doesn't matter.

In case it fails the worse part is re-use the box after I drilled it for larger hole caused by XLR jack.

Edit:
Just found it.

I used the Edcor 10k:10k in my boxes after playing around with the Triad, Xicon, and whatever else I could find to try.

I played around with RG inspired Humfree ABY circuit A LOT, built it stock, tried several transformers, sounded great for regular guitar but stuffy in the bass when I played baritone or Bass VI, got the Edcors... sounded really great!  Then I started to get crazy and built things to drive them harder to get the sound of a mic pre getting pushed hard, at one point I was using the discrete op amps from a API clone and an attenuator on the output to push them.  Sounded great, but it was kind of over complicated overkill.

Now, I just drive them with good neutral buffers that are on my board and the transformers are actually in boxes next to my amps, one has a phase switch to compensate for amp differences and they both have the fancy version of the Jack Orman pickup simulator hard wired on the secondary, one of them has a pot that approximates my tone knob to balance an amp I think is a little bright with the darker amp on the other side of the room.

Basically, I found it's best to stick with pretty much RG's thing unless you're playing baritone or bass, then stick with RG's thing and use a bigger transformer.  I personally like putting the transformers closer to the amps because using the bigger ones takes up more space on my board and there's less high frequency loss from the long cable run (in my studio, one of my amps is about 40 feet from my pedalboard), because I'd rather use tone knobs and stuff to control that a little more precisely.

The RG inspired Humfree ABY schematic http://www.geofex.com/FX_images/spltr2.gif should be a buffer (or a slight booster) in front of a transformers.
What wavley means is that I could have some problem with bass frequency with the TY-250P? Or it depends from the schematic part before the transformers.

I remind, my friend use low tuned guitars, sometime bass guitar, too.