EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Totsuken on April 05, 2024, 03:38:01 pm
-
Hi guys
I have started to repair/restore old amps as a hobby. I haven't been doing it very long so I'm cautious and have the amps isolated when working on them. Up until now I have just been using software on a tablet as a signal generator, but it would be more convenient to have a proper one on the bench ready to use. The thing I'm worried about is compromising the isolation of the amp if I connect a mains powered generator to it. Would it defeat the isolation by connecting the signal generator? I know there are ones with isolated outputs, but they tend to be highter end ones and I can't justify getting one of those.
Thanks
-
Buy a 1 to 1 isolation transformer.
-
I do have one. Do you mean to power the signal generator with it?
-
Presumably the signal generator is just for test purposes? If so I wouldn't worry. The input being connected to earth isn't going to damage it.
If you need to isolate the output of the signal generator, then use an audio isolation transformer, not a mains transformer.
-
Thanks Zero
Like I say, I'm just learning as I go. I just put signals in to test with the scope if something doesn't seem right. I don't really know if I need to isolate the output. The reason I ask is because I have watched the eevblog youtube video on the importance of having the equipment under test isolated to protect the scope from mishaps. If I connect a mains powered signal generator to the amp, won't that lead to the same situation of a possible shorting of the scope as if the amp wasn't isolated? Or am I misunderstanding that video due to my lack of knowledge?
-
Thanks Zero
Like I say, I'm just learning as I go. I just put signals in to test with the scope if something doesn't seem right. I don't really know if I need to isolate the output. The reason I ask is because I have watched the eevblog youtube video on the importance of having the equipment under test isolated to protect the scope from mishaps. If I connect a mains powered signal generator to the amp, won't that lead to the same situation of a possible shorting of the scope as if the amp wasn't isolated? Or am I misunderstanding that video due to my lack of knowledge?
Yes, connecting a mains powered signal generator or oscilloscope to an unearthed amplifier will ground it, but that's not a problem in itself.
The problem occurs when you connect an earthed signal generator to a part of a circuit is referenced to earth, yet at another voltage. The most commonly given example is the mains side of a switched mode power supply. The neutral of the mains is connected to earth, so connecting the oscilloscope causes a short circuit. This isn't the case with an isolated amplifier, which is not connected to earth, so there's no path for the return current,
-
But won't there be a return path to the amp through the generators leads, even if the amp itself is isolated? That's what I'm worried about. Sorry to keep coming back, but when I'm learning from people who know what they are talking about, I like to make sure I'm understanding them correctly.
-
A return path to earth is not a problem... The issue is mains live/neutral getting to the ground of the amp, not earth getting there.
-
Hi Mark
Yes I know, but I'm talking about the circumstances it shows in that "How NOT to blow up your Oscilloscope" video. It says if you connect the scope ground lead to live by mistake, the scope wouldn't be damaged as long as you're running the device off an isolation transformer, which I do. What I am asking is: If I connect the signal generator to the amp while testing and happened to make a mistake like that, then wouldn't there be a short back through the scope lead > earth > signal gen > amp again?
-
Theory would suggest that even though the chassis would be earth grounded, having the mains input coming through an isolation transformer would still allow you to ALSO place the ground clip somewhere within the mains side of the SMPS and make a measurement without blowing things up. HOWEVER, I would suggest never working on the hot side of a SMPS while also doing tests on the 'user' side of the supply (the audio section) and generally speaking you would never be working on both sides at the same time. Usually when an amp comes in for repair it is either the audio section or the power supply that has failed, not both. Basically the isolation transformer allows you to make measurements on the hot side of a SMPS without hurting you or your test equipment. You can however still blow up the power supply depending on where you place your ground lead. For instance, you would never make a measurement that required placing the ground clip on the gate or base of the main switch FET or Transistor. It would load the gate drive or disturb the base drive and end up killing a perfectly good working SMPS. Caution and thinking ahead are the words for the day. Cheers, enjoy your new test gear.
-
If the amplifier is isolated, then it must be powered off an isolated supply, i.e. isolation transformer, or battery.
The don't blow up your 'scope video refers to connecting the oscilloscope to a non-isolated power supply.
-
What kind of amplifier?
Many old school Audio amplifiers used inbuilt power transformers, inherently offering mains isolation.
the transformerless design common in domestic radios & TVs in some countries were often done where the cost of power transformers were regarded as an expense which could be avoided.
If the inbuilt insulation in such devices was not breached, they could arguably be regarded as "safe", but amplifiers, by the nature of their use were often "hands on" devices, offering more chances of getting "zapped" if a transformerless design was used.
They were normally more expensive than TVs & Radios, so the cost incentive did not really exist.
-
Thanks guys. None of the gear I buy use a SMPS, they're old 70's amps with transformers. And, as I say, I do use an isolation transformer with the amps anyway. But looking at the diagrams that guy shows on the video, explaining how the short can be created when the device isn't isolated, it just seemed that a short would be possible again once I connect up a signal generator to the amp. Or am I still not getting something? I just want to know if the danger is still there.
-
But looking at the diagrams that guy shows on the video, explaining how the short can be created when the device isn't isolated,
Short can be created if you connect your signal generator output to power 'live'. There is no power 'live' inside amplifier (I suppose), if you will not connect your generator output BEFORE transformer in power supply of amplifier.
-
Thanks guys. None of the gear I buy use a SMPS, they're old 70's amps with transformers. And, as I say, I do use an isolation transformer with the amps anyway. But looking at the diagrams that guy shows on the video, explaining how the short can be created when the device isn't isolated, it just seemed that a short would be possible again once I connect up a signal generator to the amp. Or am I still not getting something? I just want to know if the danger is still there.
The port you connect your signal source to has nothing to do with the Mains.
If the audio amplifiers you are working with have power transformers, the Mains connection is confined to the on/off switch & the power transformer primary.
The metal chassis, cabinet or "earthy" part of any PCBs are not connected to the Active or Neutral Mains lines, but, if they use a three conductor power cord, the chassis will probably be connected to the building's Protective Earth.
With modern devices, the special situation of trying to look at SMPS with a 'scope can give rise to the problem referred to in the video, but the DC supplies at the output of the SMPS are transformer protected by the higher frequency transformer used in SMPS.
WARNING:- Some manufacturers of 1970s/80s TVs were somewhat less than scrupulous at physically separating Mains wiring from other wiring, but electrically the isolation still exists.
"Transformerless" devices were formerly common amongst domestic radios & TV sets in some countries, including the UK, & with them, there is a good chance that the chassis may be inadvertently connected to the Active conductor, so if you connect a "ground" lead connected to the building PE as would be the case with a 'scope to that chassis, you will get a very large spark, & (if you have one), your RCD will instantly operate.
Even with such devices, however, if they had provision for an audio input from a record player (traditionally marked PU), both sides of the connection would be isolated from the chassis using large value, high voltage rated capacitors, as the wiring from the PU cartridge of record players is not in any way Mains rated, quite apart from the high likelihood of the cartridge disappearing in a puff of smoke!
-
Thanks vk6zgo
I've been re-reading all the posts and it suddenly clicked why I couldn't take in what you guys have been telling me. I was too busy thinking about what the chap in the video said, and not about what I actually do. I'm obviously not going to damage my scope with the voltages in the amps I work on.
Is my understanding that I could cause a short circuit through the signal generator correct though? If the scopes ground lead touched say a 30v or 40v rail, would this come back through the grounds of the signal generator and the amps RCA inputs, and possibly cause damage to components in the amp or signal generator?
-
An audio amplifier already has an isolation transformer powering it. Here's an example, of a basic amplifier showing the isolated power supply.
[attachimg=1]
Yes, you can cause damage by connecting the signal generator and oscilloscope's ground.common leads to different voltages, but that's true, irrespective, of whether an isolation transformer is used or not. For example, in the above, schematic, if you connected the 'scope's ground to +V and the signal generators ground to 0V it would short circuit the power supply and bad things will happen. The moral of the story is always connect the earth of the signal generator and oscilloscope to the same part of the circuit.
-
Bonju=our bravo for the fine question.
We have worked in Pro/Cineam/high end udio since 1970s, and used many odscillators, generators, and now mobile apps.
Best zero cost solution is modern mobile audio gen apps: Sine, sweep, noise, MLS, funct gen avail ijunk or android free.
But Phones without a 3.5 mm HP jack need an adapter to it.
We also used the BT speakers like JBL Go2 which has a 3.5 mm HP jack!
Setup
Phone(app)>>3.5 mm cable>>RCA >>Amp under test.
OR
Phone>>BT spkr>>3,5mm>>amp under test.
For theare EQ, we use pink npise gen apps.
Play thru system, use RTA, 1/3 Oct SA, and calibrated mic to survey the room, HT, venue.
VOILA! No transformers, mains, isoation perfect, no hum.
Occams razor: Reducto ad absurdum.
HAVE AN ABSOLUTELY AFNTASTC DAY!
jon in Paris
-
Thanks vk6zgo
I've been re-reading all the posts and it suddenly clicked why I couldn't take in what you guys have been telling me. I was too busy thinking about what the chap in the video said, and not about what I actually do. I'm obviously not going to damage my scope with the voltages in the amps I work on.
Is my understanding that I could cause a short circuit through the signal generator correct though? If the scopes ground lead touched say a 30v or 40v rail, would this come back through the grounds of the signal generator and the amps RCA inputs, and possibly cause damage to components in the amp or signal generator?
It would be unlikely to damage the signal generator, or the components used in those parts of the amplifier which fulfill its primary purpose, but it may well damage the DC supply circuitry.
It can't "come back through the grounds of the signal generator & the amps RCA inputs" because once it is shorted out, it is shorted out.
-
Brilliant Zero999.
That explains it perfectly and is exactly what I wanted to know. I obviously didn't ask the correct questions at first. I am always careful to connect probes at the appropriate places but, as much as I hate to admit it, I do make mistakes sometimes. :-[
Bonjour jonpaul
Until now I have been using that method on my tablet, but I just fancied getting a proper signal generator. But now I understand the pitfalls, I might just stick to doing it that way for the time being.
Hi vk6zgo
Sorry, I'm not sure of the proper terms to use in electronics. What I meant to ask was: would that be the path the short? Does that make sense?
Thanks for taking the time to answer my questions. I love forums like this. It's like having a load of mates that know everything!
-
I have had moments of doubt about clipping the scope probe's ground somewhere, and just simply used the DMM to measure the voltage from scope ground to that point before clipping the clip.
-
I do have one. Do you mean to power the signal generator with it?
It is usually better to galvanically isolate the device under test, so the amplifier, rather than the test instruments because of safety.
-
I have had moments of doubt about clipping the scope probe's ground somewhere, and just simply used the DMM to measure the voltage from scope ground to that point before clipping the clip.
I can't believe I never even thought about doing that.
-
If the scopes ground lead touched say a 30v or 40v rail, would this come back through the grounds of the signal generator and the amps RCA inputs, and possibly cause damage to components in the amp or signal generator?
Only if both scope and signal generator (or your amplifier) has direct connection between internal circuit ground and Mains ground wire. It could be, in theory, but almost impossible in reality.
In any case there is not advisable to connect grounds of oscilloscope (or signal generator) to anything except ground in DUT, even with isolating transformer.
You can easily create short cut without any galvanic coupling to Mains or whatever. For example: you have 2 channel oscilloscope and try to connect one channel between ground and some signal and other between +24V and another signal. This configuration will short cut 24V supply for 99.999% of oscilloscopes.
-
I have had moments of doubt about clipping the scope probe's ground somewhere, and just simply used the DMM to measure the voltage from scope ground to that point before clipping the clip.
This should be printed in 36 pt font size in bright red as a "sticky" at the top of every thread that has people questioning where they can put their scope probe.
Experienced people very seldom have a problem, but noobs seem to always have one.
-
Only if both scope and signal generator (or your amplifier) has direct connection between internal circuit ground and Mains ground wire
That's why I asked this question. I don't have a mains powered signal generator, but I was thinking of getting one.
In any case there is not advisable to connect grounds of oscilloscope (or signal generator) to anything except ground in DUT, even with isolating transformer.
No, I would never do it deliberately. I was worried about doing it through lack of knowledge, or accidentally touching the wrong place while placing the ground lead.