DIY Audio Projects Forum

Hi Everyone, Every now and then something comes along that just doesn't make sense. This is one of those. The circuit is a Groove Phono Preamp with Lundahl SUTs. It is one of the preproduction ones for Oddwatt. It is doing some very strange things and has me stumped. The sound is great, no issues there. However when you look at the signal to noise spectrums it is baffling. In MM it is not really too bad. Yes I would like all hum etc to be well below the -74 range it has, but I could live with it. When you go to LOMC things get odd. With the inputs open with no connections both channels are at about -70dbv. Not bad really. When you short out either the S/N jumps to about -63. Say what...it should go down not up. Odder still is the actual spectrum. With open connections the problem area is at 120HZ. Short the inputs and 120 drops way down and 180 and 300 pop up. Essentially it goes from mostly first harmonic to second and higher ones. I don't ever recall this behavior. Moving things around on the ground buss alters the behavior but the general increase in power supply harmonics is about the same. No arrangement makes it really all that good. Any thoughts?

The schematic is the same as several earlier ones with the exception of built in SUTs and the PCB. I attached the PCB with some notes.

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

Seems like you got an oscillator?
The transformer may have a high step-up ratio and a large inductivity and there are some stray capacitances? (this is just me thinking out loud...)
I´d try to use some small capacitors (trial and error up to some hundreds of pF) in parallel with the input and see what happens.

Miguel

I'm thining the power supply and the heater scheme if you did the ground and can't solve the noise prolbem.

Hi, I tried that and it still is behaving in an odd way. There is no oscillation at any frequency that I ca measure (near DC to over 25 meg). The SUTs are set for 16X. I have used them before without any issues. I strongly suspect some part of the physical layout is acting sort of like an antenna and causing the problems. Just putting a piece of steel near one section makes the hum and noise (especially at 180HZ) jump up several db. If the metal is more than about 5cm away it goes back to normal. It doesn't matter if the metal is grounded or not. I have revised the ground conenctions, routing and sequencing several times. Alternately grounded and not the pcb ground planes (there are 3) and so on. The preamp is pretty quiet, but if you take it out of the case it gets about 5db quieter and I would like to have that improvement. It bugs me. BTW above about 300HZ the S/N is over -90dbv.

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,

Try joining the G for each channel together then one wire from the highest gain section ground to the ground bus.

Attach a long wire between ground bus and Ground. Then get a piece of ferrite something like an old AM radio Ariel rod and start winding the ground wire round the rod one turn at a time.

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

Hi, Did something like that. I used mini inductors. Just got more hum. All things are grounded via a single connection. Some improvement, but did not eliminate the sensitivity to nearby steel. However, I discovered that shorting the inputs to the SUTs increases the hum and noise...not sure why, but putting a 100 ohm load on the input side helped a lot. I suspect that they were seeing the ground and thought there was stuff on it. Since their ground terminals were attached to the same place and the distance is really short that does not seem logical. The input to the tube is connected to the same place as are the outputs of the SUT. On the MM/MI inputs shorting does help, but doesn't get the hum and noise as low as moving the pcb away from the metal. That part just works backwards to what is expected. There is a X2 and 120R separating the metal from the circuitry as a ground break, a pair of reversed rectifiers (often seen in pro gear) was tried as well. It didn't matter in either case. My understanding of shielding etc must have a hole (my spouse says just one of many ) in it as this is not something I have run into before.

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,

Here is a thought for you.

You are detecting metal<<a metal detector.

OK so how does a metal detector work..

You have an energizer coil and a search coil. So for the metal to have an effect it must be interacting with the circuit in some way.
We know that grounding it has no effect, so the only other way it can interact is via a magnetic field.

So you must be energizing the metal with eddy currents from the transformers, and the other is picking it up like a search coil.

By loading the coils you are effecting the interaction of the circuit by effecting the magnetic field in some way.

So is the energizing field coming from the circuit? if it is can you stop this by creating a magnetic circuit around the transformers that force the field into a controlled situation or some kind of faraday cage? So OK it hums with metal in the area but what happens if you put metal into the circuit note the hum level then compare if it changes when in or out of the box. ie you make the circuit ignore the case?

Think about the car lock transmitter that is rendered inactive with a bag, or the mobile phone that stops working inside a microwave oven.

Something else worth a try for a laugh is does metal with holes in it have the same effect as solid metal?
Ie the holes create a magnetic path for the eddy currents like the rotor in a squirrel cage motor.

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

Hi, I like the analogy. It describes the behavior better than anything else I have thought of. It should not bother the MM part of the circuit as the SUTs are no connected at that point. Even disconnecting the wiring to them before the mode switch doesn't fix it. The remaining circuitry is pretty standard and has never been an issue before. Still ....I will try an external power source tomorrow and see if that helps.

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

Does this amp previously work the low noise?
Otherwise some amps are designed with flaws and it can never become low noise, so just accept it will never be less noise than a chicken.

Hi, This is a new design and being debugged. I have done a bit of research on this problem and a lot of experimenting on it. I have found two things that matter. One is relatively simple. Separate ground busses and changes to the main circuit grounding scheme. The other is black magic. Moving the power transformer around (twisting, turning, and placement) makes a huge difference. Using shields (steel in this case, but I tried aluminum and Mu metal) doesn't really matter. Copper shielding doesn't matter either. Changing to a toroid trannie or a different standard type didn't matter either. The toroid actually made it worse. Using an aluminum case instead of the steel one didn't matter either. Moving the transformer far away externally did reduce the hum and noise to about -82db but that really is not feasible in this build. Just moving it around altered the spectrum of the hum. It would change from 120HZ to 180HZ. Quite interesting to watch this happen in real time. It would also shift it from one channel to the other. My assessment on the situation is that one portion of the circuitry in the vicinity of the on board SUTs was acting as a receiver of EMF/EMI from the transformer. Even though the SUTs are Mu metal encased and were behind a steel shield. Lots of grounding there etc. So clearly a PCB redesign is needed. Such fun. I have a EMF/EMI test device on the way and I will see if it can make sense of the mess. Right now I'm ready to delete the SUTs and allow users to get their own external ones. It is what I do in my system. That way you can make things really really quiet. More to follow.

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