• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Getting voltage spikes at idle, with simple triode stages

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello everyone.

I built a simple preamp, using a single ECC88, and I'm getting voltage spikes at idle. Two common cathode stages, capacitively coupled. No bypass capacitors. As simple as it gets. Bias is fine. As designed. Supply voltage is 225 V dc.

I'm getting 20 V peaks at second stage output (anode), every 10 ms, with no signal input.

I see those same spikes (much lower amplitude, around 100 mV) at the input triode's grid. So they are just being amplified by the two stages.

I removed the triode from the socket, and I still get the spikes at the input.

Seem like charging spikes, at exactly 100 Hz. The power supply is just a bridge rectifier, reservoir cap (100 uF), and three RC stages after it. More than enough filtering. No center tap. Filament winding (no center tap either) use two 100 ohm resistors soldered to the ground bus.

Never happened to me before (or at least, never saw it)

Any ideas about where should I look? Thank you!
 

Attachments

  • spikes.png
    spikes.png
    93.8 KB · Views: 220
Disconnect the filament supply, while measuring, and observe the scope. If it stops, you need to dress the wiring differently. Maybe you could also try lifting the heater supply to about 1/6~1/4 of the supply voltage? This reverse biases the internal heater-cathode diode properties, and can be useful in some situations.
 
Disconnect the filament supply, while measuring, and observe the scope.

Just tried. It doesn't stop.

Waltube, thanks for the tips, but that is another story. The spikes are not related to the triodes.

By the way, it won't be a standalone preamp; probably the input stage of a power amplifier, using some GNFB. But that will come next.

By the way, if this info is of interest, this is how voltage varies at stage's supply voltage (yellow trace). It is not ripple. Look at the time base (250 ms). It seems ok to me.
 

Attachments

  • supply_var.png
    supply_var.png
    4.3 KB · Views: 205
Last edited:
100mV at the input is like an elephant in the room, it's huge, and the cause shouldn't be difficult to find by making some tests and manipulations.

For example, disconnect one wire of the mains input, then the other one.

Check in your house if there is a phase-control supply active: the pulses are spaced 10ms, but they alternate in polarity, meaning the raw 50Hz has something to do with it, not the full-wave rectified supply. (It could even originate from your scope in a way or another)

Etc, etc. : try to be creative
 
Ok, solved.

I'm not using the safety earth right now (just using two wires from the wall plug).
I find this safer for prototyping stages. Do you agree with this? This way the power transformer acts as an isolation transformer (no shock hazard if we accidentaly touch anything at earth potential), and it is safer for scope probing, as any point can be picked as ground for the probes, and a mistake cannot blow up our probes.

But I left the aluminium sheet where I'm building this prototype unconnected.

I was going to connect it to the safety earth later on, then to a single signal ground point, as I usually do, with good results.

Connecting the metal sheet to my signal ground at any point, eliminated the spikes.
I found this because placing a scope probe very close to the metal sheet, not touching it, shows a similar spike at the same exact moment, so that attracted my attention.

This never happened to me, because I always used safety earth from the beginning, or at least, once I started doing real tests on the circuit.

Maybe this was an easy one. Just I never faced it.


The supply voltage still swings randomly up to about 100 mV, example shown (yellow). The blue trace is the one at the input: now it is clean.
 

Attachments

  • solved.png
    solved.png
    4 KB · Views: 153
Last edited:
I supposed so. It seems small enought to be neglected (<0.05% of supply voltage).

Even though it was not the purpose of this thread, what about working without earth connection. Do any of you guys find a potential hazard on it?

Even if sounding repetitive, I forgot to say thanks again to all of you, for participating, and sharing your knowledge, whatever it is, and as always, the very helpful support I always get from the people in this great forum.
 
Do you agree with this? This way the power transformer acts as an isolation transformer (no shock hazard if we accidentaly touch anything at earth potential).

In general yes: in a test setup, it is preferable to avoid any high voltage, but that's not always possible, especially if you deal with tube circuits, but high voltages by themselves may not be too harmful, because you need to take into account the return path.
A high enough AC (and sometimes DC) source will give you a shock, even if it has no formal return path, but this does normally not have grave consequences.

Things gets really nasty when you have both a low impedance source AND a low impedance return path.


If you are a good boy, and carefully earth the 0V of your amplifier, there will be lot of points that could become letal if you touch them while grabbing a probe, touching the metal frame of an instrument, etc.

Morality, earthing can save lives (and it does), but when you have high potentials in a earthed environment, it can be very dangerous too.

In this context, having a maximum of floating elemnts is the safest option, because there will be no way to close a low impedance circuit.

Anyway, I'd recommend you use your discernment to analyze each situation and context, and chose the most appropriate configuration in each case.

Being yourself reasonably insulated is generally a good advice: a non-metallic chair on an insulating soil or matt, and if you need to ground yourself for ESD reasons, always do it via a 1Meg safety resistor, never directly.
 
But… to finish off… in anyone has forgotten the choice comment (paraphrased): “perhaps there is something generating positive- and negative-going spikes on your house power line?”

That something could be one of 3 “usual suspects”:

(1) “dimmer controls” for lights … especially attached to newer LED type bulbs
(2) LED bulbs themselves
(3) variable speed motor controllers, exp. for modern laundry washing machines.

The dimmer controls (used-to-) use Triacs to avalanche-turn-on the line voltage, synchronous with line frequency, somewhere in both the positive and negative going phase, at 100 Hz (Europe, non-North-America) or 120 Hz … North America.

Especially when dimming 'dimmable' LED bulbs, these avalanche conduction events induce fairly high charging current to the LED bulbs' AC→DC rectification front end. Naturally, synchronous with line frequency.

Motor-speed controllers (which have become all the rage in the last 7 to 10 years for heavy appliances such as variable-rate laundry washing machines), can pre-lower operating voltage with the same remarkably inexpensive (and durable) Triac setup. Same kind of spikes.

Lastly those silly little LED bulbs … in order to offer highest efficiencies, have quite the compact switching power supplies onboard. Never know what kind of spikes they'll put out. Especially the El Cheapo ones from Nameless China.

Just saying,
GoatGuy
 
In this context, having a maximum of floating elemnts is the safest option, because there will be no way to close a low impedance circuit.

This is one of the main reasons I keep the amplifier and chassis not earthed while building and testing.
Any of you guys think this is a bad idea?

The only potential hazard I can think of, is the case in which some high voltage touches the floating chassis. Using no earth, the chassis will be hot, without us noticing it.

What I did, is a first power up, with a wire connecting ground (not earth wire) to the chassis, using a current limiter such as bulb limitter. As no fuse blows, and the bulb stays off, I removed the wire.

But the floating chassis made all this nasty voltage spikes, so even if not earthing the chassis, grounding it seems mandatory.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.