EEVblog HVP70 Differential Probe

[mikeselectricstuff]

I used my first probe like this around 1988. I seem to remember it cost about A$800 and the bandwidth might have been 20MHz. In all the time since then, good differential probes have never changed much in price, so there has to be something difficult in designing and making good, safe probes.

More like just a big enough market for people to put much into R&D

[amspire]

I used my first probe like this around 1988. I seem to remember it cost about A$800 and the bandwidth might have been 20MHz. In all the time since then, good differential probes have never changed much in price, so there has to be something difficult in designing and making good, safe probes.

More like just a big enough market for people to put much into R&D

I don't know about that. If someone was selling good $50 differential probes, I think people would just buy them as a standard part of their scope kit. I just assume that it is hard to achieve a good safe $50 differential probe.

[Mechatrommer]

this is 70MHz unit allright, not 25MHz... just to be fair with other products brands fame and shame... when are we going to expect for dont turn it on, take it apart session of this thing? yet, a detailed specification is not yet know where to download...

[EEVblog]

He also was a fan of the Tektronix p6046 differential probe but commented on how tricky they were to trim properly after repair. They don't have the high voltage capability of Dave's probe but for someone looking for a budget differential probe, they can be found on eBay for < $100.

+1 , made this quote while ago when I was waiting to get that probe for years at cheap price.

Btw, I'm not sure these days you can get that P6046 at that price anymore.

I just checked. There's a few on eBay now for < $100.  This one looks good but is an auction. Opening bid $50.

That is not a high voltage probe, not the same thing.

[EEVblog]

I don't know about that. If someone was selling good $50 differential probes, I think people would just buy them as a standard part of their scope kit. I just assume that it is hard to achieve a good safe $50 differential probe.

$50 retail, very likely not easy.

[EEVblog]

I used my first probe like this around 1988. I seem to remember it cost about A$800 and the bandwidth might have been 20MHz. In all the time since then, good differential probes have never changed much in price, so there has to be something difficult in designing and making good, safe probes.

More like just a big enough market for people to put much into R&D

Likely.
I think most companies just reverse engineer the Sapphire and use that as their base circuit.

[mtdoc]

He also was a fan of the Tektronix p6046 differential probe but commented on how tricky they were to trim properly after repair. They don't have the high voltage capability of Dave's probe but for someone looking for a budget differential probe, they can be found on eBay for < $100.

+1 , made this quote while ago when I was waiting to get that probe for years at cheap price.

Btw, I'm not sure these days you can get that P6046 at that price anymore.

I just checked. There's a few on eBay now for < $100.  This one looks good but is an auction. Opening bid $50.

That is not a high voltage probe, not the same thing.

Yep, no doubt - and why I mentioned that in my post. Still a useful, low cost, low voltage differential probe.

[David Hess]

Yep, no doubt - and why I mentioned that in my post. Still a useful, low cost, low voltage differential probe.

The P6046 is very or even irreplaceably useful if you actually have an application which needs something like it.  This probe was Tektronix's longest lived product.

There is a big difference in performance between low voltage and high voltage differential probes.  The Sapphire and similar high voltage differential probes do not include a complete CMRR versus frequency or noise (or recovery time) specifications for good reason; they cannot compete with the P4046 in this respect or even the old Tektronix 7A13 and Preamble 1822 (bought by LeCroy) designs when they are used with the proper external probes.  Except for input voltage range, even an properly adjusted oscilloscope in add and invert mode can be competitive. (1)

On the other hand, they are perfectly suited for line side applications like off-line switching power supplies where a higher voltage range is required and are simpler to use and less expensive than a 7A13 or Preamble 1822.  And if capt bullshot's story about calibration is common, then most users do not understand or know how to accurately use them anyway.

As far as their seemingly high price, I suspect it is just a situation of what the limited market will bear and the design and validation is tricky for such a low volume product.

(1) It is a little eerie using an analog oscilloscope in add and invert mode while adjusting it for maximum CMRR.  You can *see* the displayed noise decrease when it is adjusted correctly.  The same goes for adjusting the external probes on a 7A13 or Preample 1822.  The movie opens on a fog shrouded hill; the fog clears as the audience yells, "Focus!"

[EEVblog]

Just discovered an embarrassing issue that will delay shipment of the fist batch.
#NotMyFault

[Jay_Diddy_B]

Hi,

Are the red and black leads, + plus and - minus mixed up? They are different than other sapphire probes.



Regards,

Jay_Diddy_B

[_Wim_]

Well you can't put a price tag on your own safety! IMHO it is one of those situations where you just have to bite the bullet. Come to think of it... I think the differential probes are among the most expensive probes I have (based on the price I paid for any particular probe).

For safety at a lower price tag you can also buy the Tektronix A6902B 2 channel isolator (http://exodus.poly.edu/~kurt/manuals/manuals/Tektronix/TEK%20A6902B%20Operator.pdf) for around 100$ at EBay. Not as high of a bandwidth as the differential probes discussed here (when taking into account the high frequency derating => max 3MHz for 500V, and max 20MHz to low voltages), but for sure a lot safer than floating the DUT or scope. CMRR is in the same league as more modern differential probes.

[amspire]

I have an old Hantek 60Mhz 2150 USB oscilloscope which I could run off a battery pack.

I am going to try out a USB isolator using the AD chip and see if I can make the Hantek run in a useful manner on USB 1.1. You can buy 1500V isolator boards on Ebay for about A$15. Also have some 100x 2500V 100MHz probes on order for about the same price. Don't have any projects planned for this - probably just trying to find a use for the Hantek scope.

http://www.ebay.com/itm/381860044223
http://www.ebay.com/itm/282303148790

The idea is to leave it floating. Not a replacement for a differential probe when there are huge slew rate, but fine for general checking of Mains potential circuits.

Even if the provided software does not work, there is a fair bit of Open source stuff that probably would work.

[_Wim_]

I have an old Hantek 60Mhz 2150 USB oscilloscope which I could run off a battery pack.

I am going to try out a USB isolator using the AD chip and see if I can make the Hantek run in a useful manner on USB 1.1. You can buy 1500V isolator boards on Ebay for about A$15. Also have some 100x 2500V 100MHz probes on order for about the same price. Don't have any projects planned for this - probably just trying to find a use for the Hantek scope.

http://www.ebay.com/itm/381860044223
http://www.ebay.com/itm/282303148790

The idea is to leave it floating. Not a replacement for a differential probe when there are huge slew rate, but fine for general checking of Mains potential circuits.

Even if the provided software does not work, there is a fair bit of Open source stuff that probably would work.

Floating a scope can be dangerous because scope housing could go to a high potential (scope was not designed to be used as floating).
For example with your proposal, when you probe with 1 input of the Hantek, the outside of the other BNC connectors can (will) go to a live voltage. Accidently touching one of these connectors poses a real danger. Also metal screw used for assembly of the housing could be dangerous. As the inputs of the Hantek are not isolated from each other, probing with 2 channels can cause dangerous short circuits.

I would also be weary to trust a chinse USB isolator (fake chips?) to protect me and my pc from high voltages. Imagen you pc becomes “live” and you want to operate your scope.

So I would recommend against your proposed approach. If it was this easy (cheap) to probe safely at high voltages, all scopes would add have this feature as standard…

[_Wim_]

Also the USB cable between Hantek and USB Isolator, and half of the USB isolator will be at high potential when probing around. USB cables are not designed for this.

[amspire]

Floating a scope can be dangerous because scope housing could go to a high potential (scope was not designed to be used as floating).
...

The case would definitely be high. I wouldn't dream of touching it if it was attached to a live HV circuit.

For example, if I ever work on developing mains switching supplies, I always now put the device under test behind a shield. I have seen enough explosively fragmented mosfets and diodes to know this is a good idea. The live USB scope would also be behind the shield.

It is not a great idea to be touching the probes of a differential probe when the circuit is live.

I would also be weary to trust a chinse USB isolator (fake chips?) to protect me and my pc from high voltages. Imagen you pc becomes “live” and you want to operate your scope.

I would test it. There are 3 parts across the PCB gap - an Optocoupler, a Capacitor and the IC. I would go to the full 1500V in my tests.  Might even try two isolators in series with the first one attached to a very heavily earthed powered hub on its output. That would be safe.

As I said, I don't have a use for this yet - just want to see if the idea works. May be I would use it more often to see differential voltages on low voltage circuits - such as monitoring the voltage across a current resistor in the positive line of a grounded low voltage power supply.

[bson]

What is the output impedance of this to feed a 1Mohm input at 70MHz?

Looks interesting though, if the price is right I'll definitely buy one to have on hand, even if I very rarely have an urge to go poking around in the mains side of supplies...  (Priced right would be, say, US$499 or less.)

[_Wim_]

Floating a scope can be dangerous because scope housing could go to a high potential (scope was not designed to be used as floating).
...

The case would definitely be high. I wouldn't dream of touching it if it was attached to a live HV circuit.

For example, if I ever work on developing mains switching supplies, I always now put the device under test behind a shield. I have seen enough explosively fragmented mosfets and diodes to know this is a good idea. The live USB scope would also be behind the shield.

It is not a great idea to be touching the probes of a differential probe when the circuit is live.

I would also be weary to trust a chinse USB isolator (fake chips?) to protect me and my pc from high voltages. Imagen you pc becomes “live” and you want to operate your scope.

I would test it. There are 3 parts across the PCB gap - an Optocoupler, a Capacitor and the IC. I would go to the full 1500V in my tests.  Might even try two isolators in series with the first one attached to a very heavily earthed powered hub on its output. That would be safe.

As I said, I don't have a use for this yet - just want to see if the idea works. May be I would use it more often to see differential voltages on low voltage circuits - such as monitoring the voltage across a current resistor in the positive line of a grounded low voltage power supply.

I just wanted to make this clear for beginners reading this that it is not intrinsically safe. Many beginners will not realize a live housing and could seriously harm themselves.
I fully agree with nctnico that you cannot put a price on safety, even (especially!) if you are a beginner/hobbyist. So when working on higher voltages, only use probes specifically designed for this purpose.  Because  I do understand that buying a couple of high voltage differential probes can be a big investment, I recommended the Tektronix isolator above with can be bought cheap. This is old technology (was probably quite expensive in its day), but safe and specifically designed for this purpose.

An even safer solution is NOT probing high voltages (mains) at all with a scope. Most info you need to know can be provided by a multimeter.

[David Hess]

What is the output impedance of this to feed a 1Mohm input at 70MHz?

It depends on the length of the unterminated cable is but probably 50 ohms.  70 MHz is just on the edge of being problem and if tested with a reference level pulse generator; there is probably a visible step in the response.

There is actually a good reason not to calibrate the output assuming a terminated cable and use a 1 Mohm input; the termination accuracy degrades the gain accuracy.

[bktemp]

I have an old Hantek 60Mhz 2150 USB oscilloscope which I could run off a battery pack.

I am going to try out a USB isolator using the AD chip and see if I can make the Hantek run in a useful manner on USB 1.1. You can buy 1500V isolator boards on Ebay for about A$15. Also have some 100x 2500V 100MHz probes on order for about the same price. Don't have any projects planned for this - probably just trying to find a use for the Hantek scope.

http://www.ebay.com/itm/381860044223

Looking at the pictures, I doubt those isolator boards are useable for anything near 1kVpeak: The DC/DC looks like a cheap one rated for only 500Vpeak isolation voltage with the primary and secondary winding wound on top of each other without rainforced insulation.

[amspire]

You are right - I hadn't looked closely. That is great - I can take it off. I have power both sides.

[capt bullshot]

If suitable components are used, then usually it is the printed circuit board material which changes (or starts out bad) screwing up the high impedance compensated dividers.  Tektronix called this "hook" and before qualifying standard fiberglass weave printed circuit board material, early on they used boards made from plastics like polysulfone (amber looking) for their high impedance attenuators and later a white board material that I have not identified yet.

Yes, that PCB material issue isn't totally unknown to me, wrong material can screw up your circuit. But I don't know anything about PCB material parameter change over time. Sometime it's better to use Teflon standoffs or similar construction techniques - way to expensive for todays world.

Jim Williams had something to say about maladjusted probe compensation:

"The translation of this statement is to hide the probe when you are not using it.  If anyone wants to borrow it, look straight at them, shrug your shoulders and say you don't know where it is.  This is decidedly dishonest, but eminently practical.  Those finding this morally questionable may wish to re-examine their attitude after producing a day's worth of worthless data with a probe that was unknowingly readjusted." - page 12 of Linear Technology application note 49)

Thanks for that quote! I remember having read this some time, and sometimes I follow his advice, like "yes, you may borrow the scope but find your own probes".

[David Hess]

If suitable components are used, then usually it is the printed circuit board material which changes (or starts out bad) screwing up the high impedance compensated dividers.  Tektronix called this "hook" and before qualifying standard fiberglass weave printed circuit board material, early on they used boards made from plastics like polysulfone (amber looking) for their high impedance attenuators and later a white board material that I have not identified yet.

Yes, that PCB material issue isn't totally unknown to me, wrong material can screw up your circuit. But I don't know anything about PCB material parameter change over time. Sometime it's better to use Teflon standoffs or similar construction techniques - way to expensive for today's world.

The Tektronix article discusses problems with epoxy resins so I wonder if it would change over time if improperly cured.  Humidity is also a problem with some board materials.  I suspect I have run across problems with hook in connection with dual slope integrating analog to digital converters.

I think the lesson is to either use the right board material to start with or include a charge test circuit as described in the article for quality control.

[thm_w]

I don't think people realise how much these probes currently sell for. You can't even buy a cheapie ebay or Aliexpress 25MHz job for $200, let alone a quality 70MHz one.

https://www.aliexpress.com/item/New-Professional-Differential-probes-DC-25Mhz-ADP25-Max-Voltage-1300V-1000V-RMS/946453698.html
https://www.globalmediapro.com/dp/A01LT0/Pintek-DP-25-Differential-Probe-25MHz-1000V/

$183 25MHz
$290 65MHz

Teardown: https://www.eevblog.com/forum/testgear/teardown-pintek-dp-25-differentail-probe/

[technogeeky]

Just discovered an embarrassing issue that will delay shipment of the fist batch.
#NotMyFault

Hi,

Are the red and black leads, + plus and - minus mixed up? They are different than other sapphire probes.



Regards,

Jay_Diddy_B

That would be pretty embarrassing...

[alanb]

Just discovered an embarrassing issue that will delay shipment of the fist batch.
#NotMyFault

If you've a batch with the leads the wrong way round I would take one off your hands for £50!