I use 60/110W toroidal output transformers of my own design. Very happy with them.
3.9K with CFB - Cathode feedback; Ultra-Linear Tap 40% (w/o CFB) - unit can be used as UL, CFB, and UL + CFB.
Nominal power - 60W (optimal flux density B);
Max power - 110W (max flux density for toroid B = 1.75T);
Frequency response - 10Hz - 65 KHz (practical limit determined by resonant frequency);
Measurements with CFB:
Primary inductance - 20Hz/1V - 359H, 100Hz/1V - 116H;
Leakage inductance (secondary shorted) - 1KHz - 9.88mH, 10KHz - 9.41mH, w/o CFB - 8.83/8.41mH (1/10 KHz);
Primary resistance (DC) - 82.11 Ohm;
Secondary resistance, DC (8 Ohm terminal) - 0.28 Ohm;
Secondary resistance, DC (4 Ohm terminal) - 0.18 Ohm;
External diameter - 14 cm
Weight - 4.6 Kg
Temperature raise at 60W/20Hz (1 hour) - 7 - 10 degree C (measured with infrared thermometer).
This toroid can sustain 7mA of output tube current imbalance (e.g. 43 mA for one tube and 57 mA for another), without any distortions seen on oscilloscope at 20Hz/high power.
Since transformer is quite large, upper frequency limit is determined by resonant frequency caused by leakage inductance/stray capacitance, and not by leakage inductance alone.
Eliminating CFB would allow to further enhance characteristics and reduce cost, but I like how CFB amplifier sounds
Transformer is designed with my own CAD software BA-TrafoWiz.
3.9K with CFB - Cathode feedback; Ultra-Linear Tap 40% (w/o CFB) - unit can be used as UL, CFB, and UL + CFB.
Nominal power - 60W (optimal flux density B);
Max power - 110W (max flux density for toroid B = 1.75T);
Frequency response - 10Hz - 65 KHz (practical limit determined by resonant frequency);
Measurements with CFB:
Primary inductance - 20Hz/1V - 359H, 100Hz/1V - 116H;
Leakage inductance (secondary shorted) - 1KHz - 9.88mH, 10KHz - 9.41mH, w/o CFB - 8.83/8.41mH (1/10 KHz);
Primary resistance (DC) - 82.11 Ohm;
Secondary resistance, DC (8 Ohm terminal) - 0.28 Ohm;
Secondary resistance, DC (4 Ohm terminal) - 0.18 Ohm;
External diameter - 14 cm
Weight - 4.6 Kg
Temperature raise at 60W/20Hz (1 hour) - 7 - 10 degree C (measured with infrared thermometer).
This toroid can sustain 7mA of output tube current imbalance (e.g. 43 mA for one tube and 57 mA for another), without any distortions seen on oscilloscope at 20Hz/high power.
Since transformer is quite large, upper frequency limit is determined by resonant frequency caused by leakage inductance/stray capacitance, and not by leakage inductance alone.
Eliminating CFB would allow to further enhance characteristics and reduce cost, but I like how CFB amplifier sounds
Transformer is designed with my own CAD software BA-TrafoWiz.
Toroidal OPTs are used in the Velleman K4040 kit, which can work very well after mods described elsewhere in this forum. Also, Plitron sells toroidal output transformers. IMHO, the pros of toroids are compactness and low susceptibility to external fields, e.g. from the power transformer. The cons are low tolerance to imbalance of the PP output stage (that is, earlier saturation with DC current) and the price.
Toroidal Transformers Amplimo - Ringkerntrafo.nl
sells them also, output and powersupply transformers.
sells them also, output and powersupply transformers.
"I have been playing with maybe using a toroidal power transformer. So the question is has anyone used toroidal output transformers?"
If you are thinking of using a toroidal power transformer as an output transformer, there are some issues to consider. The usual dual 120V primary (at 50/60 Hz rating) will become rather voltage limited by magnetic saturation at say 20 or 30 Hz. (120V rating at 60 Hz --> 40 V rating at 20 Hz)
The primary inductance of toroidals is generally quite good compared to similar EI xfmrs, but if you scale the size/watts upward enough, you can run into a problem with the primary inductance. Doubling the core area allows half the turns for the 120V rating, this causes the inductance to drop by 1/2 (2x area but 1/2 turns squared).
The distributed capacitance of power toroids is generally quite high, like 100 to 1000X or more that of comparable audio xfmrs. ("comparable" is key here since the power xfmr generally has 10X fewer turns than the audio designed OT) Special winding techniques are used in the toroid OTs designed for audio to control this.
If you are thinking of using a toroidal power transformer as an output transformer, there are some issues to consider. The usual dual 120V primary (at 50/60 Hz rating) will become rather voltage limited by magnetic saturation at say 20 or 30 Hz. (120V rating at 60 Hz --> 40 V rating at 20 Hz)
The primary inductance of toroidals is generally quite good compared to similar EI xfmrs, but if you scale the size/watts upward enough, you can run into a problem with the primary inductance. Doubling the core area allows half the turns for the 120V rating, this causes the inductance to drop by 1/2 (2x area but 1/2 turns squared).
The distributed capacitance of power toroids is generally quite high, like 100 to 1000X or more that of comparable audio xfmrs. ("comparable" is key here since the power xfmr generally has 10X fewer turns than the audio designed OT) Special winding techniques are used in the toroid OTs designed for audio to control this.
Last edited:
The secret to success with using toroidal power transformers is to use low output impedance tubes and aim for Plate to Plate impedances of between 1K6 and 3K. This will help to overcome the high interwinding capacitance and give plenty of bite on the inductance. I have always used about 100VA toroidals for any output wattage from 5W -20W. They have all shown responses from 10hz out to 50khz. Connecting the primaries in one direction will produce a high frequency resonance at about 60khz (which is very audible) and this will send the transformer into saturation with significant impacts on response below 100hz. Simply reversing the primary polarity eliminates this issue.
The power toroidals I have used only tolerate DC imbalances of a few mA and so I have used a CCS biasing arrangement with cathode to cathode capacitance bypassing as suggested to me by Brian Beck. This restricts operation to Class A but makes for a differential output stage which sounds excellent.
They are my rules for creating a cheap but excellent sounding toroidal based output stage.
Cheers
Shoog
The power toroidals I have used only tolerate DC imbalances of a few mA and so I have used a CCS biasing arrangement with cathode to cathode capacitance bypassing as suggested to me by Brian Beck. This restricts operation to Class A but makes for a differential output stage which sounds excellent.
They are my rules for creating a cheap but excellent sounding toroidal based output stage.
Cheers
Shoog
For properly constructed toroid this is not a problem (see my post above).
Price? For large quantity may be does matter, for few custom build pcs price will be about the same.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.