PG506 calibrator generator voltage error


Gif Sim
 

good morning to the whole group
I ask for help because I'm not understanding anything anymore ....
I have a pg506a that makes a strange defect
I turn on the pg506a and connect an oscilloscope (I have tried with various oscilloscopes) to the amplitude output in standard mode
the output voltages are not correct (they are higher) in any position of the amplitude switch
if I extract and rotate the pot variable out nothing changes and the display remains fixed on a value
if I connect a multimeter or a battery powered oscilloscope (ths730) to the amplitude output, everything works correctly
the voltages are correct and the pot variable out works as expected
It seems to be a ground fault
I connected the pg506a to the extender from the tm504 to measure the supply voltages and they are all correct (I followed the service manual)
I hope I have explained myself as well as possible
ask if anyone have any suggestions to be able to understand the problem
thanks
Simon


Rick
 

Simon, the PG506's require a 50 ohm termination on the scope. Without it the voltages will be erroneously high. On mine it even adds a lot of distortion to the pulse. Do you have a 50 ohm terminator on the line?

Rick


Gif Sim
 

hi rick
50 ohm termination is used in fast rise outputs
I don't use fast rise outputs
I connect to the standard amplitude output which supports 50 ohm and 1Mohm
if you look at the amplitude standard output female bnc you will see that the ground is isolated from the pg506a frame with a white plastic insulator
The amplitude standard output has a floating power supply to ground
I connect the oscilloscope input (which is powered at 220v mains voltage) to the standard voltage output of the pg506a....
the mass of the oscilloscope bnc is connected to the network ground
when I connect it to the output of the pg506a the central variable out potentiometer no longer has any effect .... the display shows a fixed number and the output voltage is wrong it is higher ....... if I connect a standard output to a battery powered multimeter or oscilloscope everything works and the voltages on the output bnc are right ... if I turn the central potentiometer the voltage varies and the display works perfectly (indicating the% of error) .........
something happens inside the pg506a !!!
but I can't understand what!
with the service manual I checked the power supplies and some diodes, zener, transistors, connections ....... everything seems okk ....

I do not know what to think !!! I'd start crying .. hey


Gif Sim
 

thanks rick for your interest ..... I'm going to sleep ... it's 3:00 am .... maybe tomorrow I'll do other tests and measurements and I can tell you more ..... bye


Eric
 

Simon,
The calibrated output and the high amplitude output (same BNC) are cal in to 1Meg Ohm. The fast rise are cal in to 50 Ohms. Have you checked the calibration on the PG 506? If I recall correctly it is done at DC. There is a switch inside the calibrator to stop the squairwave so you can calibrate it with a DMM. If you do that you may need to match the 1Meg input of the DMM there was a way to do that in the service manual. Also if I remember correctly from when I calibrated mine there was some things about the contacts in the relays being a soar spot on these but I could be mistaken on that.

Eric

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Rick
Sent: Saturday, December 26, 2020 7:43 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] PG506 calibrator generator voltage error

Simon, the PG506's require a 50 ohm termination on the scope. Without it the voltages will be erroneously high. On mine it even adds a lot of distortion to the pulse. Do you have a 50 ohm terminator on the line?

Rick


 

I have repaired a number of PG506's and these are the common problems:

Intermittent operation of the flat white relays at the back of the instrument
On older instruments, there are two small metal-cased relays near the front. These switch in a low-pass filter for some standard-amplitude settings to reduce noise from the switching power supply.
Cam switch contacts. To get to these, it is necessary to remove the DVM daughter board
The standard amplitude cal trimpots R340 and R205. I replace the single-turn pots with multi-turn trimmers

To clarify the 50-ohm vs 1 megohm load question, the Standard Amplitude output is intended to drive a 1 megohm load at all ranges. At 10 V and below, it can drive 50 ohms, in which case the output should be exactly half the 1 megohm load value. The High-Amplitude output can drive either a 1 meg or 50-ohm load. The output to a 50-ohm load will be about a tenth of the 1-meg value.

The fast-rise outputs must drive a 50-ohm load.


Bob Haas


Gif Sim
 

hi eric
yess..I did as you wrote ... the pg506a used it long ago and it was perfectly calibrated and working (I calibrated some 2465b and 2467b)

the strange thing is that if I connect an oscilloscope with 1Mohm input, battery powered, therefore not connected to the mains electricity, to the amplitude standard output, the levels are correct ... the variable out potentiometer works and the display indicates the value set with the pot ....

but if I connect an oscilloscope (such as a tds 784d) with 1Mohm input, powered by the mains electricity, the output voltage is wrong (for example if I set 2 v to the output or about 2.4 v) ... .the variable out potentiometer has no effect on the output voltage and the display indicates a fixed value

my calibrator is the PG506A and one of the latest revisions of the boards there are no relays you find in the first PG506

I checked q325-q365-q326-q320-q280-q290-q245-q255-q270-q190 by removing them from board A1 and trying them with a component test they are all okk...

i checked all diodes and zener on board A1 and they are all okk
now i am trying and replacing some electrolytic and tantalum capacitors on board A1 (but so far i haven't found any wrong ones)

Eric can you do me a favor? ... with the PG506 off, set to standard ampl and the selector on 2volt, if you connect the cables of a multimeter between the mass of the standard out female bnc and the aluminum frame of the PG506, what a resistance read the multimeter?

thanks eric and thanks to all those who can give any suggestion to understand the problem!
simon


Eric
 

Hey Simon,
Measured with a Keithly DMM6500 standard output jack to the aluminum frame. In a frame and out of a frame.
In frame, resistance that rapidly went up to overflow. I believe a cap was charging.
Out of a frame, Same behavior.

Once charged infinite resistance.

Eric

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Gif Sim via groups.io
Sent: Sunday, December 27, 2020 2:10 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] PG506 calibrator generator voltage error

hi eric
yess..I did as you wrote ... the pg506a used it long ago and it was perfectly calibrated and working (I calibrated some 2465b and 2467b)

the strange thing is that if I connect an oscilloscope with 1Mohm input, battery powered, therefore not connected to the mains electricity, to the amplitude standard output, the levels are correct ... the variable out potentiometer works and the display indicates the value set with the pot ....

but if I connect an oscilloscope (such as a tds 784d) with 1Mohm input, powered by the mains electricity, the output voltage is wrong (for example if I set 2 v to the output or about 2.4 v) ... .the variable out potentiometer has no effect on the output voltage and the display indicates a fixed value

my calibrator is the PG506A and one of the latest revisions of the boards there are no relays you find in the first PG506

I checked q325-q365-q326-q320-q280-q290-q245-q255-q270-q190 by removing them from board A1 and trying them with a component test they are all okk...

i checked all diodes and zener on board A1 and they are all okk now i am trying and replacing some electrolytic and tantalum capacitors on board A1 (but so far i haven't found any wrong ones)

Eric can you do me a favor? ... with the PG506 off, set to standard ampl and the selector on 2volt, if you connect the cables of a multimeter between the mass of the standard out female bnc and the aluminum frame of the PG506, what a resistance read the multimeter?

thanks eric and thanks to all those who can give any suggestion to understand the problem!
simon


Gif Sim
 

hi bob
thanks for the tips I'll go check it out immediately!
as i wrote to eric i have a pg506a there are no relays ...
the thing I don't understand is that if I connect a battery powered oscilloscope everything works correctly ... all controls and values are normal
i noticed that q290 q190 and r190 are very very hot but taking them apart and trying them they are okk ... also vr210 is ok the 9v are present
I noticed that when I connect an oscilloscope that can be tds784d or 2465b, then powered by the mains
the voltage +120 volts (which I measure on cr291) rises to +125 volts .....
I'm going crazy !!!
thank you all
simon


Gif Sim
 

thanks eric !
what main frame do you use to power the pg506? (I think one of the tm500 series)
can you do the same test with the pg506 inserted in the tm500 and the multimeter connected between the mass of the standard out female bnc and the mass present in the power plug of the tm500?
I think the problem is here .... in my I measure 10kohm ....
thanks
simon


Jean-Paul
 

Are you sure of the grounds and mains power connector, could it be a grounding or loop issue?

jon


Harvey White
 

What it sounds like is that something (likely a capacitor) *may* be shorted.  That one capacitor that charges might be bad, so I'd wonder if the sneak path for grounding through the power line is not shorting out something, like for instance, the 120 adjust control?  Now I don't mean that the control is shorted out, but the ground that used to be ground isn't any more, and that throws off the circuit.

Harvey

On 12/27/2020 2:37 PM, Gif Sim via groups.io wrote:
hi bob
thanks for the tips I'll go check it out immediately!
as i wrote to eric i have a pg506a there are no relays ...
the thing I don't understand is that if I connect a battery powered oscilloscope everything works correctly ... all controls and values are normal
i noticed that q290 q190 and r190 are very very hot but taking them apart and trying them they are okk ... also vr210 is ok the 9v are present
I noticed that when I connect an oscilloscope that can be tds784d or 2465b, then powered by the mains
the voltage +120 volts (which I measure on cr291) rises to +125 volts .....
I'm going crazy !!!
thank you all
simon





Gif Sim
 

hi jon
yes ... that's one of the first things I checked
the grounding of the instruments and the bench are ok ..
i also opened the tm504 to check the ground and a couple of resistors connecting the ground ... all ok
the pg506 has two floating masses .... at this moment I have the pg506 on the bench
if I measure the resistance between the two masses I read 10-11 kohm
I think it's not correct ... Eric has just measured an infinite value between the two masses on a working pg 506 .....
I performed the measurement between the two masses on C100 ....


Gif Sim
 

hi harvey
that's exactly what I think too !!
in fact if I connect an oscilloscope isolated from the mains power supply (battery powered) the pg506 works correctly ....
only that inside the pg506 it's a nice mess !! .....
simon


Harvey White
 

I just leafed through the schematic, and don't see a capacitor between the two grounds (there are, but they're very small values, like 0.01 uf).  What I'd do is to disconnect the relay board, which should isolate the power supply.  I'd check to see if the resistance you see goes away (I'd expect that the normal behavior is very high resistance between grounds.  There *is* a diode (CR67, I think) between the 5 volt (referred to ground) and isolated ground.  I'd wonder about that.  )

Then I'd check between the two grounds again and see which section has the odd reading.

If you want, I could check one of mine to see if it's behaving itself.

Harvey

On 12/27/2020 3:26 PM, Gif Sim via groups.io wrote:
hi harvey
that's exactly what I think too !!
in fact if I connect an oscilloscope isolated from the mains power supply (battery powered) the pg506 works correctly ....
only that inside the pg506 it's a nice mess !! .....
simon





Gif Sim
 

harvey
it's a PG 506A has no relay .....
I removed and measured C100 .... I measure 10-11kohm ... instead it should be a common 100 nF ceramic....
replaced with a new capacitor ... now between the two masses I measure about 2.5 Mohm ..... it seems okkkk !!!
give me some time ... I reassemble the pg 506A .... and I can tell you if it works .... I cross my fingers !!! .........


Gif Sim
 

yess ... yesss .... it was c100 now the pg506a works correctly !!!!

never happened that a ceramic did so '... usually electrolytic and tantalum give problems
I hope it will be useful to someone .... before writing here I searched a lot on the net but I didn't find much ...

is it normal for Q290 and Q190 to be very very hot?
is there a way to make the output less noisy?
in standard amplitude on 1 Mohm oscilloscope I measure 2 mV P-P
is this a normal value?

THANKS GUYS !!!!!!!!!..........IT'S ALSO THANKS TO YOU !!!!!!!!!

Simon


Tom Lee
 

Ceramic caps rarely become defective from simple age, but can be damaged, as can any other component. For example, if a previous user connected the PG506 to something whose common-mode voltage was very greatly different, that cap could've suffered.

You haven't described the noise in any specific detail, so your question can't be answered in any specific detail. Generally speaking, verify that your supplies are clean. And try to provide more information if you want to get more helpful responses. Is the noise manifesting as timing jitter? Amplitude noise on the flat portions? Does the noise appear to have structure? etc. The more you can tell us, the more we can tell you. And in the process of finding more yourself, you will frequently be led to the problem's cause on your own.

Congratulations on getting this far.  I'm sure you'll find that plug-in a very useful tool!

-- Cheers,
Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 12/27/2020 14:29, Gif Sim via groups.io wrote:
yess ... yesss .... it was c100 now the pg506a works correctly !!!!

never happened that a ceramic did so '... usually electrolytic and tantalum give problems
I hope it will be useful to someone .... before writing here I searched a lot on the net but I didn't find much ...

is it normal for Q290 and Q190 to be very very hot?
is there a way to make the output less noisy?
in standard amplitude on 1 Mohm oscilloscope I measure 2 mV P-P
is this a normal value?

THANKS GUYS !!!!!!!!!..........IT'S ALSO THANKS TO YOU !!!!!!!!!

Simon




Gif Sim
 

hi thomas
Thank you for your contribution
you're right ..... I was very superficial in describing the "noise" I see with the scope
tomorrow I will upload some screenshots of the oscilloscope and some info on the settings of the pg508a and tds784d
here in Italy it is midnight .... I was late tonight ..... I ate now .... I go to sleep
the pg506a is really a very useful plug-in .... I use it a lot .... I have the whole series pg506A-tg501A-sg503-sg504
bye
simon


Harvey White
 

If you have all of those, and more than one 7000 series mainframe, you may want to look at the standardizer plugins for the 7000 series.

They put out a standard set of waveforms/voltages and replace a plugin.  They allow you to standardize the input sensitivity of a 7000 series scope (they have gain controls on the input channels *in the mainframe* which need to be adjusted *if* you have more than one frame and *if* you want the plugins to be calibrated once they're plugged in.  If you only have one 7000 series frame, then in a sense, you calibrate the plugins to that so they're all correct, and you have no other frame to transplant the plugins to.

Be aware that the standardizers come in different varieties (check last digits in the type name) for different bandwidth scopes.  They can be used in either the V or H channels (using one in each with the right settings gives you a grid).  One other ability they have is to be able to convert a variable amplitude (as in.... not constant amplitude) signal to a constant amplitude signal.  I suspect that the different plugins have different frequency ranges for the sinewave input.  If you have  a constant amplitude signal generator, you don't need that feature.  Setting the input signal amplitude can be a bit tricky, though.

A lower bandwidth plugin can do most of what you'd want for the frame, but you can't check the bandwidth of the frame completely with a lower bandwidth plugin.

Harvey

On 12/27/2020 7:24 PM, Gif Sim via groups.io wrote:
hi thomas
Thank you for your contribution
you're right ..... I was very superficial in describing the "noise" I see with the scope
tomorrow I will upload some screenshots of the oscilloscope and some info on the settings of the pg508a and tds784d
here in Italy it is midnight .... I was late tonight ..... I ate now .... I go to sleep
the pg506a is really a very useful plug-in .... I use it a lot .... I have the whole series pg506A-tg501A-sg503-sg504
bye
simon