Connecting negative supply oscillator to scope
This is an oscillator that has a negative supply -15v.
An easy way to power it will be to connect a lab psu positive terminal to the ground of the circuit, and the PSU GND to the -15v point of the circuit.
However, if I do this, how can I connect my scope to the circuit without damaging it, since the PSU and the scope might share the same GND.
Before solving the scope issue, I have a feeling that the circuit is not oscillating in this configuration.
I think something is wrong, or is missing from the schematic.
Oh but it does.
It has just being redrawn from this pdf.
Did you build the circuit?
I still have doubts that oscillate..
inside/outside DC block.
Replying to this question, I've had accidental short circuits when I'm not careful. My scope is grounded to mains, and occasionally I experimented with circuitry powered from mains. I always tried to remember to connect the live AC so it never had a direct path to scope ground.
However I sometimes forgot what I was doing, and a short circuit happened. It resulted in something getting burned up. (I never bothered to get an isolation transformer.)
Is your 15V coming from a transformer-based supply? I have a transformer in my homemade low voltage power supply. It never allows a short circuit to mains AC, either through my scope or any other equipment.
Yes it is a lab psu, transformer based. It has a GND connection and a negative connection which are connected together with a piece of metla (removable).
I have not built it.
It is presented in another page as well http://www.sp-qrp.pl/modules.php?nam...howpage&pid=25 Why do you think it won't work?
What is this?
From the link you sent, the oscillator in Fig. 1 is a well known source-coupled wide band FET oscillator which works fine, but the one using negative supply I don't think can meet the necessary feedback to get in oscillation.
Anyway, would take few minutes to build the circuit and prove that I am wrong.
Can you move the metal link to between GND and positive? That would solve your problem.
Brian.
in general, one NEVER leaves the earth ground connected to the negative output. let it float.
In general, if I have my circuits under test connected to the scope, I do not need to connect the ground to the circuit to work, but just the positive. The gnd goes through the scope, so there is a connection actually.
So If I disconnect this, I effectively have two grounds, one for the scope and one isolated for the PSU?
In other words the PSU ground will be isolated from the resd of the grounds in my lab?
I do not understand why he uses negative voltage. Is that only to have the LC grounded at one end, so that hand effects are minimized?
Ground connection is meant to be a safety feature; all your instruments (the metal body) are supposed to be connected to the ground. It is not supposed to be used as a return path for the power.
The ground marked on the circuit diagrams are just common tie points. Not to be confused with the earth ground...
My scope (and other equipment) does not have a floating negative. The negative is internally conencted to the ground (I think).
But my PSU has one. Is that safe to disconenct this link from the PSU and connect the negative (floating) point to the rest of the circuit ground?
Can you please give me a clear view of how to connect this circuit that requires a negative PSU, to my scope, provided that the negative of the PSU will be connected to the -V of the circuit?
Any 'negative' should be internal to the scope, all you have is a ground clip and probe. The scope's ground clip might be connected to the Earth pin on the power plug if it has one and that creates the risk of shorting out the power supply if it also has a similar connection.
The way to do it is to isolate the Earth connection on the power supply (fit the link back afterwards if you want to) then connect the ground clip to the ground point on the oscillator, in this case the positive line. Use the probe as normal. If you use DC signal coupling, the scope trace will go down from 0V position but still be accurate, if you use AC coupling, the trace will float according to the 'Y position' control.
It is normal to make lab PSUs with no permanent Earth connection but give the user the option of linking one or the other output if wanted. It makes the PSU far more versatile, especially if more than one PSU is 'stacked' to give more output voltage or linked together to produce more than one output voltage. For example, if you had two PSUs and wanted to produce +12V, 0V and -12V, you would link two single 0V-12V suplies in series and use the center connection as 0V, it both PSUs had permanent grounds on their negative sides, one of them would be shorted out.
You can also eliminate the Earth connection on the oscilloscope but that can cause more problems that it's worth. It means the scope ground clip and probe could potentially float at high voltage and damage the circuit as it is connected. It's usually best to keep the scope Earthed and let the PSU float.
Brian.
The way to measure a floating voltage on a scope is to use two probes; leave the ground clips unconnected. on the display select A-B mode and you will get the potential difference between the two probe points...
The very high common mode rejection of the input amplifiers for the scope makes this an easy task. Well, you have certainly seen this trick done with op-amps.
I'm assuming there should still be a safety Earth connection but c_mitra is right about using differential measurements. I'm not sure if it would be applicable in the case of high frequency measurements unless special care was taken to match and double screen the scope cables though.
Brian.
Ok so to summarize:
1.Disconnect the link on the PSU, so that I have separate GND and negative on the PSU.
2.Connect the PSU positive to the circuit ground.
3.Connect the PSU negative to the circuit -VCC
4.Connect the RF output of the circuit to the plus terminal (central conductor) of the input of the scope.
5.Connect the circuit ground (which is now connected the positive terminal of the PSU), to the ground terminal of the input of the scope.
Is this correct?
Yes, and optionally, link the Earth connection on the PSU to the positive output connection. If the scope has it's own Earth connection it shouldn't be necessary but if it hasn't there would be no Earth anywhere so for safety, add the link at the PSU. There are no strict rules on this, if safety is more of an issue, connect the Earth but it can cause loop problems if there is more than one Earth point in a circuit and there is any potential difference between them.
Be careful when referring to 'positive and 'negative' as oscilloscope inputs. We know what you mean but technically they are 'ground' and 'probe' as oscilloscopes work equally well at measuring positive or negative on the probe tip.
From an electrical perspective, it doesn't make any difference at all whether the positive rail or negative rail or neither of them is connected to Earth but you have to be sure that if there is more than one Earth connection, they are all wired to the same polarity.
Brian.
That is pervectly clear!
Thanks Brian and all of you in this thread
You solved the scope test setup, now let us know if the circuit is oscillating.