Suggestions for DC current probes

[georges80]

^ yes.

cheers,
george.

[David Hess]

Yeah, the TM5xx is the power supply unit. There's ones with 2 slots and ones with 1 slot. You only need the 1 slot (smaller), but often the 2 slots are all you can find.

Is a drawer installed into the right side of the TM502A for holding the probe?

[timb]

I think I figured it out -

The TM501 and 503 are the powered chassis for the AM5xx amplifiers? Is that right?

Exactly, but they're not just for the AM5xx amplifiers! The TM 5xx are Tek's universal power mainframes used from the mid-70's until the late 80's. They produced a few dozen separate instruments for them. Everything from DMMs and power supplies to frequency counters, function generators, amplifiers and even tiny oscilloscopes!

The last digit in the model number of the TM (which I assume stands for Tek Mainframe) series designates how many slots it contains. So a TM 501 is a single slot unit and a TM 506 is a six slot unit.

The various plugin designations have similar names:

AM5xx = AMplifier
PG5xx = Pulse Generator (Square Waves)
SG5xx = Sine Generator
FG5xx = Function Generator
TG5xx = Time Mark Generator
DM5xx = Digital Multimeter
PS5xx = Power Supply

Plus several other types.

They're surprisingly useful mainframes as you can often get the plugins pretty cheap. I just picked up a SG 504 250MHz to 1GHz Leveled Sinewave generator for $100!

Just make sure you don't accidentally buy a TM 5000 series mainframe or plugin, as they're not backwards compatible with the 500 series.

The 500 series mainframes are pretty simple. It's essentially a huge transformer with a few dedicated windings (providing low voltage AC) and (rectified and filtered) 12VDC for each slot on the mainframe along with a high current (rectified and filtered) 32VDC bus shared among all the slots. There's also one NPN and PNP transistor for each slot (bolted to the metal frame for heatsinking).

If you buy a mainframe from eBay, I'd recommend replacing the caps. There's a couple of 5000uF ones plus a big 20,000uF one (if I remember correctly and at least for the three slot unit). Don't try desoldering the wires that go from the board to the transformed; instead just unbolt the transformer and unscrew the board and remove them together. Once out of the mainframe there's plenty of room to easily unsolder the caps. They use those old 4 terminal Sprague caps, however the hole spacing was standard, so you can replace them with your choice of snap-in style caps.

[georges80]

I've put up the PDF manuals for the A6302 and the AM503B on dropbox ->

https://www.dropbox.com/sh/px9ko9dh10375r3/AADyieBVn4oMcGY3PwvlrDeGa?dl=0

cheers,
george.

[rx8pilot]

Thank you George!

The notes on these Tek probes is very helpful - I was not looking at them before. It looks like they are the best value out there while meeting the technical needs. Fantastic.

[woodchips]

Silly question, why do you need DC measurements?

A low frequency DC&AC probes are cheap, as are AC only probes, when I was looking around that is what I decided on, but so far yet to use the DC&AC probe and wondering what I am missing.

[joeqsmith]

Silly question, why do you need DC measurements?

A low frequency DC&AC probes are cheap, as are AC only probes, when I was looking around that is what I decided on, but so far yet to use the DC&AC probe and wondering what I am missing.

Looking at biased AC signals.   Output from a rectifier, etc.

[Someone]

Silly question, why do you need DC measurements?

A low frequency DC&AC probes are cheap, as are AC only probes, when I was looking around that is what I decided on, but so far yet to use the DC&AC probe and wondering what I am missing.

Looking at biased AC signals.   Output from a rectifier, etc.

Yes, you can hook up two probes to the same signal to look at the DC-xxxkHz and xxxHz-xxxMHz separately. Its unusual to need the DC and MHz signal accurately combined simultaneously and I've worked on many power supplies with an AC scope probe and a DC clamp meter.

[georges80]

DC and AC simultaneous measurements allows you to monitor things like LED current regulated switchers that transition from DC to various PWM outputs at various times. Many of my LED drivers will go from DC output to PWM output and transition between at various output levels (e.g. strobe modes).

Being able to monitor the entire current profile on a single screen and capture it for documentation and also for average current measurements is something that a single AC/DC capable current probing setup allows you to do very easily. i.e. we're talking about a dynamic situation that goes from DC to DC/PWM to DC.

Anyhow, given the demographics of this Forum are you actually hinting that a certain piece of test equipment isn't 'needed' 

cheers,
george.

[rx8pilot]

Silly question, why do you need DC measurements?

A low frequency DC&AC probes are cheap, as are AC only probes, when I was looking around that is what I decided on, but so far yet to use the DC&AC probe and wondering what I am missing.

In my case - I almost never work on AC power at all. My products are DC from beginning to end.
I have to document inrush current and design the eFuse circuits to match. This process requires the edge rate of the current waveform to be known. Switching losses, transient response,  RDS(on), efficiency, are other applications. The shape of the curve provides insight into various problems in PCB layout or components. When trying to build DC power electronics - especially high performance - DC currents move fast. I am always trying to improve switching speeds by lowering parasitic inductance and faster switching devices. The edge rates are getting really fast.

I am also buying high-performance differential amplifiers to measure current sense resistors in places that are impossible to get these relatively big probes.

[nctnico]

I'm not sure where your product goes but fast edge rates seem unwanted to me for EMI/EMC problems. A fast edge can couple into circuit easely causing all kinds of mayhem.

[rx8pilot]

That is the challenge. The slow edges are less efficient and the fast edges make a ton of noise. The eGAN FETS are on the table now and offer the possibility of super fast switching with nearly no dead time. The transistor transits the linear region very fast and takes advantage of the low Rds(ON) even with fast switching frequencies. The next challenge is to deal with the resulting EMI mess it can make  - as you pointed out. It is a new area for me, but I think it is a good pursuit. Smaller and more efficient power conversion and management.

A less challenging fast edge rate problem I have to deal with is step currents and inrush created by the loads. I have to be able to know what the loads need and how to make the supply, protection, and monitoring systems stable in the face of fast changing currents. In those cases, I am doing what I can to slow down the slew rates to something more manageable.

[Someone]

DC and AC simultaneous measurements allows you to monitor things like LED current regulated switchers that transition from DC to various PWM outputs at various times. Many of my LED drivers will go from DC output to PWM output and transition between at various output levels (e.g. strobe modes).

Being able to monitor the entire current profile on a single screen and capture it for documentation and also for average current measurements is something that a single AC/DC capable current probing setup allows you to do very easily. i.e. we're talking about a dynamic situation that goes from DC to DC/PWM to DC.

Except here the OP is asking for corner case specifications while wanting it cheap, so alternatives are being suggested. You can measure edge rates of switching supplies with an AC probe, the content below 1kHz is not needed. At the same time you can determine or measure the DC separately. Its convenient to do it all at once but look at the OPs requirements they originally stated, there are precious few probes with both the bandwidth and a 30A DC continuous rating.

[David Hess]

Yes, you can hook up two probes to the same signal to look at the DC-xxxkHz and xxxHz-xxxMHz separately. Its unusual to need the DC and MHz signal accurately combined simultaneously and I've worked on many power supplies with an AC scope probe and a DC clamp meter.

This will almost always *not* work.  Wide bandwidth AC only probes saturate at low DC currents making them useless if there is DC bias.  You can use them to measure in series with a capacitor but not a switch or diode.  A Tektronix P6021 saturates at 0.5 amps and a P6022 saturates at 0.2 amps.

Wide bandwidth AC/DC probes have a counter winding which suppresses DC saturation. The counter winding is also how they are able to measure DC current.

[rx8pilot]

Except here the OP is asking for corner case specifications while wanting it cheap, so alternatives are being suggested.

To be clear, I am not looking for cheap at the expense of function. I am looking for the function at the lowest practical price. In all reality, the 30Mhz is likely a solid 5x more than a will see any time soon - but I prefer the headroom in bandwidth. The max RMS DC I would expect on a normal day is 15A, so I wanted to double that and still have some transient headroom for the day when I want to push the DUT to it's destruction. Some days I would measure steady state DC current, others will be looking at how fast can the circuit switch before it becomes an unwanted radio station.

The biggest lesson so far is that there are indeed precious few probes in this range to pick from. Thankful to learn of the Tek DC options, as they seem to fit the bill and are not very expensive. Really thankful for the discussion - thank you all.

[Someone]

Yes, you can hook up two probes to the same signal to look at the DC-xxxkHz and xxxHz-xxxMHz separately. Its unusual to need the DC and MHz signal accurately combined simultaneously and I've worked on many power supplies with an AC scope probe and a DC clamp meter.

This will almost always *not* work.  Wide bandwidth AC only probes saturate at low DC currents making them useless if there is DC bias.  You can use them to measure in series with a capacitor but not a switch or diode.  A Tektronix P6021 saturates at 0.5 amps and a P6022 saturates at 0.2 amps.

Wide bandwidth AC/DC probes have a counter winding which suppresses DC saturation. The counter winding is also how they are able to measure DC current.

There are more probes outside of the world of Tek/Hioki, AC coupled loop probes based on Rogowski coils are very versatile and are "immune" to DC bias:
http://www.pemuk.com/products/cwt-current-probe/cwt-ultra-mini.aspx

[rx8pilot]

http://www.pemuk.com/products/cwt-current-probe/cwt-ultra-mini.aspx

Pretty interesting product.


For the Tek AM503B amplifiers, I have started making efforts to buy a pair of them. One thing I noticed on at least two options is that the OVERLOAD light is lit with nothing connected. Is that an indication of a problem or does it go away when a proper probe is connected?

[David Hess]

There are more probes outside of the world of Tek/Hioki, AC coupled loop probes based on Rogowski coils are very versatile and are "immune" to DC bias:
http://www.pemuk.com/products/cwt-current-probe/cwt-ultra-mini.aspx

Those are neat but give up a lot of sensitivity and low frequency performance.  For an equivalent low frequency cutoff, they have a noise level of 750 milliamps peak-to-peak versus 750 microamps peak-to-peak or about a thousand times worse than a P6021.

It looks like the most sensitive one is good for 30 amp measurements with a useful resolution down to 100 milliamps and 20 millamps with averaging which is fine for most applications.

They are pretty expensive compared to a wide bandwidth AC or AC+DC current probe which can be found used.

[georges80]

http://www.pemuk.com/products/cwt-current-probe/cwt-ultra-mini.aspx

Pretty interesting product.


For the Tek AM503B amplifiers, I have started making efforts to buy a pair of them. One thing I noticed on at least two options is that the OVERLOAD light is lit with nothing connected. Is that an indication of a problem or does it go away when a proper probe is connected?

My AM503B WITHOUT a probe connected, flashes several LEDs on power up but NEITHER the Overload or Probe Open or Degauss lights are lit after power on.

So, no idea why they would be on in the units you are looking at to purchase...

cheers,
george.

[rx8pilot]

Well - I just purchased 2x of the AM503B's in a TM502A enclosure along with 1x of the A6302 probes. Still looking for a second probe but I am thankful to have these units on the way. As always, eBay is a cross-your-fingers endeavor. All of it is able to be returned if there is a problem.

[TheSteve]

Looks like some successful best offers - hope it all works exactly as it should.

[timb]

Well - I just purchased 2x of the AM503B's in a TM502A enclosure along with 1x of the A6302 probes. Still looking for a second probe but I am thankful to have these units on the way. As always, eBay is a cross-your-fingers endeavor. All of it is able to be returned if there is a problem.

AM503 is one of my saved searches, so earlier while browsing eBay I came across (what I think) was that very set. I got up to let the dog out, came back and the item was sold; I thought "Must have been rx8!"

I was going to PM you a link to it, but I guess you were a step ahead of me. Looked to be in nice shape, too.

[rx8pilot]

$475 for the amplifiers and another $175 for the probe. Not bad at all!

Sent from my horrible mobile....

[georges80]

Good job, I'm sure you'll be happy with their performance (assuming all is up to spec).

When you turn on the AM503B it will require you to degauss the probe. Then a few minutes later it'll light up again (as it warms up and decides you need to degauss). It'll do this a couple or so times, usually after 20 minutes or so it'll be stable for hours on end without any more requests to degauss. Just an idiosyncrasy as it warms up on each power up. It's usable once degaussed, you don't have to wait 20 mins etc to actually use it or get readings.

I use it much more often than some of my other probes (active, differential, etc) that's for sure. It can be very enlightening to see some of the current spikes that various bits of gear go through on power up - lets you see things that you would never otherwise notice... might cause lack of sleep in some cases

cheers,
george.

[lukier]

Great score! Could you do a mini-teardown of AM503B when you receive it? I have the oldest crusty AM503 with A6302 and it is not that great, the amplifier seems noisy (pick up) and drifty. I've read somewhere that modern AM503 (A & B) are much better in that regard and I have AM503A service manual, I've seen the photo of the insides somewhere on the internet and I'm curious how much different AM503B is.