DIY Audio Projects Forum

Hi, As stated in various posts....I buy transformers and do not make them. This is so it is easy to duplicate the circuit. A consequence of this is they have certain electrical characteristics that I can not change and need to take into account. All transformers can resonate in a circuit under various circumstances. These amps with pure resistive loads are stable and do not generally have issues when the NFB is disabled. The more the load deviates from resistive the greater the chance of problems. Without NFB the amp respond (depending on what other components and the actual way it is built) to well into the self resonant range of the transformers. I have measured several with significant (not full power) output as high as 85-90KHZ. When the external load is not resistive it will introduce both impedance variations and reactive variations. These will be reflected back into the amp and can cause interaction with the transformer parameters (and indeed other components). When I observed this behavior (on a scope) I introduced the negative feed back loop shown in the schematics. It is tailored to reduce the gain at above audio band frequencies. In doing so it prevents the interaction between transformer and load from mis-behaving. Within the audio band it has virtually no effect other than stabilizing gain by a small amount. Frequency response is not altered. With most speaker systems it is possible to defeat the NFB and be fine, some particularly those with several crossovers can be suspect. The biggest issue is that the amp may mis-behave at above audio frequencies and you not know it. It will just sound different than it should. In severe cases I would expect some sort of distortion at high power in the above 10KHZ range. My test data (I really abused some of the early amps) seemed to indicate overshoot and ringing above that range with highly inductive and highly capacitive loads that simulated crossovers above about 8KHZ when trying to deliver essentially full power at those frequencies. A much more severe test than most and really outside the parameters of any sort of reasonable use. I went so far as to test full power square waves in one set of amps at 5KHZ. Abuse for sure. Unfortunately I lost most of the scope shots from back then to a "ransomware" virus that got past my checkers.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Thank you Suncalc. As much as I understand the design of FB circuit is important. How do we calculate the capacitor and resistor values in FB circuit?

Bruce, as much as I understand I don't have to change anything in NFB, even if I use a different OPT since NFB you designed covers audio band and takes care of above audio band freq.

Hi, The easiest way to calculate the NFB is to not calculate it but determine it on a scope. I have not found calculating it to be very accurate as there are many unpredictable variables. Far easier IMO to feed in a low power square wave at various frequencies above 5K HZ (I use 10K) and see what it looks like. Use a resistive load. I like to start with the resistor. BTW the ratio of the cathode resistors in the driver stage can be changed as well. But the range I like there is between about 1:5 and 1:10. Adjust the NFB resistor to have about 3 db effect on the signal output. You can use sine waves for this part. Then look at the square wave and see if it shows overshoot and/or ringing. Starting with small value capacitors (33pf, 50pf, 100pf 220pf and so on) place them across the resistor until the wave form is optimal. It may not be possible to get it perfect. A lot depends on the actual circuit components and especially the output transformer. Usually there will be a value that works well. If you find you are going much over about 500 pf then you may have to decrease the resistor in size and thus drop the gain a little more and start over with the capacitors. Usually 3 db is enough. What you are doing is forming a high pass filter with a first order slope. The corner frequency is somewhere below the range of the resonance point and when placed inside a NFB loop will reduce the gain at that frequency sufficiently to insure stability.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Thank you Bruce, nothing is better than experience for sure. Do you know primary number of turns and secondary number of turns (8 ohms and 4 ohms) of Edcor OPTs for KT120 monoblocks?

Hi, Pretty easy to calculate....20.9 for 8 ohms / 3500 ohms, 29.6 for 4 ohms / 3500 ohms

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Thank you Bruce. There are some voltage check inlets (to measure cathode voltage when balancing tubes) in the back of your kits . What is its exact name?

They are just the test points for the 1 ohm cathode resistors. Plug in a 0-200mv meter to set the balance at idle for the power tubes. It is hard to see but the adjust control is near them on most of the amps. Just a user friendly thing. I hate to have to take something apart to make a simple adjustment.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Where did you get them?

Hi, There are two ways to do this. One just use a $10 DVM. They always have a low range suitable. Get two if you want to watch how one goes up as the other down. What I did was get two inexpensive panel meters (Parts Express has some as does nearly any parts house and ebay). They are 9 volt battery operated. Each needs its own battery to keep the cathode circuits separated. Put both in a plastic enclosure and attach test leads to each. Preset the range if needed on the meters. With a DPST or DPDT switch turn the power on and off. Mine are now 5 years old and the batteries are still fine.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Hi Bruce,
Have you tried Jupiter HT Beeswax Paper as coupling caps on the signal path before? If you did, how do they perform in KT120 oddblocks? Pros and cons when comparing with Audyn True Copper caps. I'm asking because I'm leaning to Jupiter side a little bit on these days.

Can Jupiter beeswax caps endure the heat produced by K120 oddblocks?