Transformers can be used that way, such as on an isolated test bench, but that's not primarily why we use them in audio equipment. The transformer is not there to stop you receiving a shock from the audio power supply, it's to stop you receiving a shock from the mains supply. The mains supply can kill you, but the audio power supply usually cannot, as it has high source impedance, fuses, safety features, low voltages (possibly), and so on.
Yes, the voltage is low, but why risk yourself when you had already purchased an isolate transformer? Wouldn't it be much better to use proper filtering technique to reject common mode and differential noise without discarding the isolation safety feature?
For example, the circuit attached below would filtering out most of the noises from the main supply without destroying the isolation feature of the transformer.
Attachments
I understand that the risk is low. but for me, it just seemed such a waste of having an isolated transformer and not utilizing its feature. Might as well purchase a single winding auto transformer at a cheaper price.
The voltage is so low that the risk is zero. I was joking when I talked about anecdotal evidence of risk from 6.3V. You don't need isolation as a safety feature. You do need grounding (or other DC reference) for two reasons:pha0001 said:
1. the valve needs it - it should not have a floating electrode
2. you need it as a safety feature in case the transformer develops an interwinding short
Stop worrying about something that isn't a problem. Instead, start dealing with things which do need to be done. Learn about basic circuits, then later you can add a few necessary refinements. Don't sprinkle your circuits with unnecessary 'refinements' while missing glaring errors such as floating heater circuits.
That only eliminates common-mode noise, over a limited frequency range. Rectifier spikes may be partly differential mode.
Ohm law: V=IR, you have a voltage source across a resistance medium, there will be a current flow. One milliamp across the heart is enough to cause fibrillation. Thus, believed what you want but I wouldn't call a low voltage is a zero risk.
Firstly, I know my circuit theory just fine. There is no need of attacking other people on the forum. I am only here to share my opinion of what I think is the right thing to do.
Secondly. Why wouldn't a floating heater supply work? It is only floating with respect to earth. If the circuit is designed properly, there should be a common virtual earth where all circuitry is referencing to. Thus. it is not floating internally.
If what you are saying is true, then when working on valve amp, a main isolated safety bench transformer should not be used, because everything will be floating. Your logic just doesn't add up.
Last edited:
CM filtering is what I normally use for a heater supply although the 10nF are often replaced with a 500 ohm hum-pot connected to the HV elevation point. The pot is set for AC cancelling, however the AC current is still polluting the ground as one can't get it completely silent with low PSRR stages. How far goes this 'AC pollution' as the operation of the amplifier depends on ground level for several references, not at the least signal ground.
Believe me, 6.3V is zero risk.pha0001 said:
By 'floating' I mean with respect to the circuit ground (or valve cathodes). It may float with respect to safety ground - that is a separate issue. So just to clarify: the heater circuit should have a DC path to the valve cathodes. I'm sorry if I didn't make that clear.
If we have been arguing at cross-purposes, then I apologise.
He doesn't have to believe you, he can plug in the values of voltage (6.3) and resistance (>100,000 ohm) into Ohm's Law and see how tiny the current is. As an experiment, touch the ends of a 9V battery.
Where 6.3V can be a safety hazard is accidental contact with rings, bracelets, or other jewelry between the two ends of the supply. Unlikely to kill, but it's very easy to severely injure. No metal jewelry while working on any live circuit!
It won't save you. If you let the secondary system float there's no way knowing what voltage it will drift to. If you then touch it while at the same time connected to a metal grounded part (or bare feet in the floor) you'd be shocked all right!
Grounding the secondary system will at least limit the voltages to the design values. Know your enemy!
Jan
Not completely off topic 
is this issue I had last week with elevating the heaters.
This is my PS:
standing current is inbetween 20mA and 30mA.
There was an interesting but eventually very annoing resonance going on at circa 70Hz, interacting with certain toms and the kick. It drove me silly and at last I discarded the complete circuit to build a simple GC with another tube model, which (how else) suffered from the same disease. Couple of minutes ago I found that 'the lift' caused the resonance when connected before the choke. After the choke it's gone.
Before I monitored frequency response with tone generator and (high impedance) AC meter but could find no trace other than a very small (1mV) shift in level at circa 50Hz... Weird isn't it?
is this issue I had last week with elevating the heaters.This is my PS:
standing current is inbetween 20mA and 30mA.
There was an interesting but eventually very annoing resonance going on at circa 70Hz, interacting with certain toms and the kick. It drove me silly and at last I discarded the complete circuit to build a simple GC with another tube model, which (how else) suffered from the same disease. Couple of minutes ago I found that 'the lift' caused the resonance when connected before the choke. After the choke it's gone.
Before I monitored frequency response with tone generator and (high impedance) AC meter but could find no trace other than a very small (1mV) shift in level at circa 50Hz... Weird isn't it?
Because the heater supply will float up and down at high-voltage, mains frequency, due to the leakage currents flowing through the primary-to-secondary capacitance. This will result in almighty hum. Try it.
Last edited:
I'm not so sure that is always going to be valid. If the isolation transformer has sufficient performance wrt capacitive coupling from earth screening, then the floating supply is just floating, with no potential relative to earth until you touch it and you're grounded - but that doesn't form a loop.
If using just a normal transformer, then I agree that capacitive coupling to mains side and then earth will form a loop, and depending on leakage levels you may well get a good tingle.
I need to amend that comment me thinks
An isolated transformer winding, with only capacitive coupling to an earth screen, may cause capacitive current flow through an earthed person when that person touches a secondary circuit powered from that winding. The touch current level depends on which part of the secondary circuit is touched - and the windings voltage and capacitive coupling to the earth screen.
Standards are quite onerous for floating secondaries, especially if the winding has sufficient voltage to kill you, as the standard has to manage the worst-case scenario of where you touch the secondary circuit.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.