Floating OPT oscilloscope measuraments - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 7th December 2012, 10:52 PM   #1
rrrs is offline rrrs  United States
diyAudio Member
 
Join Date: Aug 2005
Default Floating OPT oscilloscope measuraments

Hi, sorry for posting this if it has been already answered elsewhere, but was not able to find straight answer in this forum.

I have a SE amp with floating OPT (secondary not grounded); this is technique recommended by Lundahl to overcome a HF roll off due to capacitive coupling between OPT primary and secondary, especially common to high impedance primary OPT's; I am using Lundahl LL1691B, 20K:8ohm.

I got a scope and would like to use it to measure a frequency response and actually compare grounded to floating OPT, but am not sure about correct way to take the measurements.
Since I do not have differential probes, if I take a measurement across the speaker terminals I will be effectively grounding the transformer trough the ground clip on the probe.
I could float my amp power supply when doing the measurement (disconnect PS from the amp chassis ground) and use battery powered (not grounded) signal generator; would that give me same result across the secondary as if secondary is floating (since in this case primary would be floating in respect to secondary)?
Other popular option would be to float the scope (disconnect the ground); of course, keeping in mind ungrounded scope could be very dangerous!; and measure across the OPT secondary. Here I am wondering about the effect of internal scope capacitance that could influence HF response?

Is there a correct way to measure floating OPT using regular scope without differential probes?
  Reply With Quote
Old 8th December 2012, 12:44 AM   #2
expert in tautology
diyAudio Member
 
bear's Avatar
 
Join Date: Apr 2002
Location: New York State USA
Why are you using such a high ratio?
Seems unusual, and to imply a very high B+??

Anyhow, I would simply use a small isolation transformer to "float" the scope. No third wire ground. (this is useful in the USA and North America where there is a 120v - neutral - ground "three wire" AC mains set up).

My scope is always isolated, because I don't want to put the ground lead inside a chassis and have the magic smoke exit.

Otoh, the sig generator, and other stuff is grounded unless I intentionally plug it into the isolated AC xfmr feed.

The sig generator you are using is connected to the input, so that shouldn't matter unless ur trying to float everything...

The scope represents a very high impedance so it should not load the output.

Btw, merely lifting the ground pin *may not* be sufficient to lift the scope's chassis above ground.

_-_-bear
__________________
_-_-bear
http://www.bearlabs.com -- Btw, I don't actually know anything, FYI -- [...2SJ74 Toshiba bogus asian parts - beware! ]
  Reply With Quote
Old 8th December 2012, 02:19 AM   #3
diyAudio Member
 
jazbo8's Avatar
 
Join Date: Jan 2011
Location: In Transient
Quote:
Originally Posted by rrrs View Post
Since I do not have differential probes, if I take a measurement across the speaker terminals I will be effectively grounding the transformer trough the ground clip on the probe.

Is there a correct way to measure floating OPT using regular scope without differential probes?
Yes, if you have a dual trace scope and A-B mode (or A+B with B inverted). Easier than floating the whole scope .

Jaz
  Reply With Quote
Old 8th December 2012, 03:57 AM   #4
diyAudio Member
 
Chris Hornbeck's Avatar
 
Join Date: May 2011
Location: Little Rock
It's potentially dangerous to float the output transformer's secondary. Insulation failure will put the speaker lines at B+. Definitely not recommended.

All good fortune,
Chris
  Reply With Quote
Old 8th December 2012, 05:10 AM   #5
diyAudio Member
 
jazbo8's Avatar
 
Join Date: Jan 2011
Location: In Transient
Agreed. Found a good post by Miles on another forum:

"It's necessary to ground one side of the secondary to provide a voltage reference for gNFB. If you don't, then there is no reference between the gNFB summing node and AC ground.

It's also a safety precaution. If the secondary is left floating, and there is a short circuit between the primary of the OPT and the secondary (let's say you forgot to connect the speeks and cranked it to 11 when you didn't hear anything. The resulting inductive kick back could cause the insulation between the primary and secondary to fail) a floating secondary becomes several hundred volts hot. That won't affect the operation, and you may not find out about it for years. That is, until you come into contact between a hot speaker while touching something grounded and get a really nasty surprise. If the secondary is grounded, then a primary-to-secondary short circuit blows the fuse. (You do fuse the primary side of your PTX, right?) That is also why it's a really bad idea to solve the problem of a three pin plug and a two slot socket with a pair of dykes to snip off the third wire safety ground. Or to build any VT equipment that doesn't have a third wire safety, and let's not even consider a design that doesn't have a PTX and runs directly off the AC mains. They didn't call such equipment "suicide boxes" for nothing.

If you float the secondary, it's essential to make certain that you can't contact a potentially hot contact, such as the speaker binding posts. VT equipment (and some solid state as well) operate at potentially lethal voltages. You want to keep those voltages where they belong, and not on the chassis, volume controls, tone controls, or anything else you might touch while the amp is working."
  Reply With Quote
Old 8th December 2012, 10:18 AM   #6
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
Adding speaker leads and a speaker to the secondary will add capacitive coupling to ground, as will attaching even an isolated scope. Big lumps of metal have capacitance as the rest of the universe acts as the second capacitor plate (and the Earth is much nearer than most of the rest of the universe).
  Reply With Quote
Old 8th December 2012, 07:28 PM   #7
rrrs is offline rrrs  United States
diyAudio Member
 
Join Date: Aug 2005
Hi, thanks to all for your comments; I agree that I would also feel better to have OPT secondary grounded; and yes, of course, my chassis is grounded and there is a fuse on the power transformer primary.
It is interesting that this method of not grounding the secondary of OPT is actually coming up a lot, especially with this high impedance transformers.
On some of the specs for frequency response Lundahl actually puts a note “secondary winding not grounded”; apparently this clearly improves HF response.
Other schematics using same or similar OPT spec a high value resistor (ex. 33K) from secondary to ground; see attached.
The fact is that Lundahl OPT seems to be built extremely well, with 4K insulation spec between primary and secondary.

For those reasons I really wanted to measure frequency response to see the actual difference as would prefer to have secondary grounded as long as my HF does not suffer too much.

Considering post of DF96, it is probably the case that even if I do float the scope to take the measurements for flouting vs. grounded secondary actual scope mass would still add capacitive coupling to ground and not show actual “floating” situation, especially at HF where I am looking to find difference.

I will try a suggestion from Jaz and measure response at both speaker terminals relative to ground and then subtract one from each other.

What is the general opinion when taking Amp measurements; should the scope probe ground clip be connected to PS/signal star ground or to the chassis ground?
I got a 5ohm resistor between PS/signal ground to chassis/safety earth wire attachment.
Attached Images
File Type: jpg SE211-6HS5-PSUincl.jpg (37.8 KB, 222 views)
  Reply With Quote
Old 9th December 2012, 04:32 AM   #8
diyAudio Member
 
Sch3mat1c's Avatar
 
Join Date: Jan 2003
Location: Milwaukee, WI
Send a message via ICQ to Sch3mat1c Send a message via AIM to Sch3mat1c
Floating the scope (on an isolation transformer, or using its internal isolation if so equipped) isn't a great idea because, face it, if it worked in the first place, we wouldn't be having this discussion: you already stated your OPT is "isolated"! We need something even more isolated-ier. A hunky power transformer isn't going to do much better than the OPT already (to be fair, it might, because a bank-wound type has lower capacitance than an interleaved shell-type OPT -- but the thing is, we don't *know* by how much).

Now, the problem is, the common mode current isn't very high (it's coming from a ~kohms primary, over maybe a few hundred pF), so it'll take a ridiculous inductance to "float" the measurement with a common mode filter. Even a pile of coax cable, threaded through high permeability toroids, will only net you an impedance in the kohms in the tens of kHz range. If the current is coming from the primary winding, its impedance will be on the order of kohms, so you need an impedance of 10s of kohms or higher.

The differential measurement using probes is handy (a 10Mohm, couple pF probe is hard to beat!), but only within the common mode range of your oscilloscope -- if the voltage turns out to be the full ~hundreds at the plates, you'll have a hard time viewing the couple of volts for the test, in lieu of a proper diff probe. Obviously, this depends on the OPT ratio. A nice low impedance OPT, maybe on a quad sweep amp or something like that, might be a piece of cake; a 20kohm OPT on a high voltage SET will have worse luck.

But all is not lost. You could take a very small transformer, which definitely has less capacitance than the OPT -- and which can be characterized -- and use that to couple the secondary voltage to your scope. It can even be some random ferrite transformer, as long as the inductance is enough for the end of the spectrum you'll be testing -- over 10kHz needs only 1mH, which is easily under a hundred turns, even on a small core. If you don't mind the turns ratio, you could easily use a transformer pulled from a switching power supply: the primary, which has to handle 300V in its intended application, will do a fine job with a few volts at relatively low frequencies.

Even more abstract methods could be used. For starters, you can measure the common-mode voltage simply by measuring the OPT secondary voltage to ground. This will be very susceptible to 60Hz hum (especially if hands are in proximity), but might still be a useful measurement.
You might infer the effect of grounding by measuring the current over a resistor to ground, or by creating a capacitance divider between OPT capacitance and a capacitor to ground. By varying the shunt capacitance, you can observe how the peaks change at the point of connection.

If you have an open OPT construction, you can try threading a turn through the core and measuring the induced voltage that way. This will give you a relative measure of frequency response, while doing a sweep or FFT or whatever, if not the actual gain. This is like having a separate feedback winding, which was sometimes done.

Tim
__________________
Seven Transistor Labs
Projects and Resources
  Reply With Quote
Old 10th December 2012, 06:42 PM   #9
rrrs is offline rrrs  United States
diyAudio Member
 
Join Date: Aug 2005
Thanks Tim, this ware really interesting suggestions!
Yes, differential measurement was hard to really do precise as AC voltage to ground on "non-grounded" OPT secondary ware over 200Vpp, so trying to measure difference across the secondary at various frequency proved to be difficult.
I tried method you suggested using additional transformer; I had small speaker impedance matching 8ohm:1.2ohm transformer which I connected across the OPT, in parallel with my "dummy" load, and then took measurements across the small transformer secondary. Here it was interesting that I got same -1db (at 13K) and -3db (at 19K) points whether OPT secondary was grounded or not!
This was surprising as measured capacitance between OPT primary and secondary is high at 7400pF, so grounding secondary should show faster HF roll-off.
I will do acoustical measurement using actual speaker and microphone with good HF extension and compare result of grounded vs. not grounded secondary; this should show the real effect at least in relative response.
Will report my findings once I get this done.
  Reply With Quote
Old 10th December 2012, 08:38 PM   #10
expert in tautology
diyAudio Member
 
bear's Avatar
 
Join Date: Apr 2002
Location: New York State USA
If the capacitance of the scope when isolated significantly effects the HF response of the "8ohm" secondary that is "floating", you'd best forget about whatever that response appears to be, because in the real world it just doesn't exist.

No doubt that differential active probes will have the least loading, but this is a low Z circuit and we're not measuring up to RF frequencies... so I'm of the view that the iso tranny works fine for this.

rrrs, you won't be likely to see the effect of the grounded vs ungrounded by measuring a speaker, because the frequencies effected are above 20Khz generally speaking.

_-_-bear
__________________
_-_-bear
http://www.bearlabs.com -- Btw, I don't actually know anything, FYI -- [...2SJ74 Toshiba bogus asian parts - beware! ]
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Measuring floating voltages with oscilloscope akis Equipment & Tools 58 17th May 2012 12:40 PM
Floating Cascode Stee Solid State 1 14th February 2010 11:30 AM
Floating the heaters YoungFred Tubes / Valves 2 16th September 2006 10:59 AM
Floating voltages eeka chu Solid State 12 29th January 2004 03:57 PM
Floating voltages eeka chu Parts 2 27th January 2004 09:10 AM


New To Site? Need Help?

All times are GMT. The time now is 08:52 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2