EEVblog Electronics Community Forum

EEVblog => Other Blogs => Topic started by: johnlsenchak on February 29, 2020, 02:59:37 am

Title: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: johnlsenchak on February 29, 2020, 02:59:37 am


Service manual  for those playing along  at home  !
http://literature.cdn.keysight.com/litweb/pdf/00419-90004.pdf (http://literature.cdn.keysight.com/litweb/pdf/00419-90004.pdf)


]"GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
•Feb 27, 2020
Mr Carlson's Lab
Lets make this Hewlett Packard, HP 419A DC Null Voltmeter work again! Very useful information regarding the chopper circuit, and how to make it dependable. Also in this video, Mr. C shares his custom designed circuit to replace the original neon chopper board.
https://www.youtube.com/watch?v=vrmwql2msbU (https://www.youtube.com/watch?v=vrmwql2msbU)

Posted  on  02/28/2019
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: rastro on February 29, 2020, 05:36:05 pm
I'm not sure why Mr. Carlson doesn't mount the LED's on the other side of the PCB. (edited)

Just realized he did do that on the installed version...
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: oPossum on February 29, 2020, 05:43:48 pm
He explained that in the video. The side of the board with the LEDs is painted black so there will be not reflections from the PCB.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Dr. Frank on March 01, 2020, 06:38:31 pm
A pity, that he as the "Tube Guy", has absolutely no clue about what he's doing with all the neons.. he obviously never heard about the Dark Effect of neons, and he did not understand, nor did he reverse engineer this neon circuit, i.e. the A3 board.
He did not measure any characteristics of all of his neons, either, like break-down and maintaining voltage. That circuit is similar or identical to a neon ring counter, so it divides the 330Hz by two, which makes these neons illuminate alternatively. 

If he would have understood, how this circuit works, he probably would have selected A3R1, A3R2 and A3C1 and maybe the driving voltage fitting to his neons, as this is the real reason, why hp at that time sold complete boards only.

Btw.: Any critical or negative comments on his YT video will be 'moderated' by him, i.e deleted.

Frank
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: johnlsenchak on March 01, 2020, 07:07:04 pm
A pity, that he as the "Tube Guy", has absolutely no clue about what he's doing with all the neons.. he obviously never heard about the Dark Effect of neons, and he did not understand, nor did he reverse engineer this neon circuit, i.e. the A3 board.
He did not measure any characteristics of all of his neons, either, like break-down and maintaining voltage. That circuit is similar or identical to a neon ring counter, so it divides the 330Hz by two, which makes these neons illuminate alternatively. 

If he would have understood, how this circuit works, he probably would have selected A3R1, A3R2 and A3C1 and maybe the driving voltage fitting to his neons, as this is the real reason, why hp at that time sold complete boards only.


Frank


I agree  with you  Mr Carson  real  didn't explain  why neon  bulbs   where chosen to   create  the oscillator circuit verses  using  germanium transistors which  where more common  when  that unit  was built 

We need  Outback  Dave to explain this  one  8)
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Dr. Frank on March 01, 2020, 08:04:18 pm


I agree  with you  Mr Carson  real  didn't explain  why neon  bulbs   where chosen to   create  the oscillator circuit verses  using  germanium transistors which  where more common  when  that unit  was built 

We need  Outback  Dave to explain this  one  8)

What?  :-//

These neon bulbs do not make the oscillator. (RTFM, page 4-3, paragraph 4-39 ff. )

It's a a blocking oscillator made up by two silicon transistors A2Q2, A2Q3, 2N3053, and transformer T2. These 330Hz are then divided by two, as the neons ignite alternatively.
So hp achieved a dead time between the two chopping phases of half the 330Hz period, i.e. about 0.75ms.
Very complicated design.
The Fluke 845AR/AB simply switches both phases seamlessly, and it works also, and that circuit is stable under all variations of neon parameters, only the Dark Effect reduction is also needed.

Frank
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: johnlsenchak on March 01, 2020, 08:49:56 pm


I agree  with you  Mr Carson  real  didn't explain  why neon  bulbs   where chosen to   create  the oscillator circuit verses  using  germanium transistors which  where more common  when  that unit  was built 

We need  Outback  Dave to explain this  one  8)



These neon bulbs do not make the oscillator. (RTFM, page 4-3, paragraph 4-39 ff. )

It's a a blocking oscillator made up by two silicon transistors A2Q2, A2Q3, 2N3053, and transformer T2. These 330Hz are then divided by two, as the neons ignite alternatively.
So hp achieved a dead time between the two chopping phases of half the 330Hz period, i.e. about 0.75ms.
Very complicated design.
The Fluke 845AR/AB simply switches both phases seamlessly, and it works also, and that circuit is stable under all variations of neon parameters, only the Dark Effect reduction is also needed.

Frank

Thanks  for  explaining  that  circuit  topology  to me, now it  makes more sense  when  Mr  Carson  was  showing the  square  waves on the  scope 8)

Much  appreciated
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: rastro on March 02, 2020, 08:21:14 pm

Is there any functional reason(s) for sticking with the neon bulbs considering the LED PCB seems to work?
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Kleinstein on March 02, 2020, 09:06:37 pm
At the time the meter was build, LED performance was not very good. So neons were a good option at the time.

This is a kind of antique meter though still good performance - so one may want to keep it close to the original.
Performance wise the LEDs are definitely better than the old neon solution. The LED solution is new replacement from Mr Carlson - AFAIK not available from HP/Keysight as a repair / upgrade part.

The only really special part is the LDR chopper - these are no longer made. For most use they are replaces by FET choppers, but they may still have some values for very low bias.

If it would be for best performance one would also replace the amplifiers, but this would no be the old meter anymore.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: gridleak on March 25, 2020, 05:10:31 am
Hope I can join this for some troubleshooting of my HP419A which still has neons installed. I added the following mod https://dabbledoo.weebly.com/hp419a.html (https://dabbledoo.weebly.com/hp419a.html). This mod makes strike and sustain voltages non-critical. I did make sure to install chopper qualified neons. The meter has been performing fine for years with the mod. Now, the meter is only behaving on 3uv through 1v ranges. From 3v - 1000 the meter wanders into the weeds either not reaching full scale on those ranges, or going off scale. The response to intermediate scale voltages is non-linear on 3v and up. The one common factor is that under 3v range the closed loop gain is positive, at 3v range, and up, closed loop gain flips negative. Can this info be used by members of this group to diagnose? I realize the easiest answer would be "your mod stopped working"... but then why are the 3uv to 1v ranges still spot on?
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Dr. Frank on March 25, 2020, 07:41:24 am
I can't fully understand the reason for this phase inversion at the high voltage ranges, as you suppose.
maybe it has something to do with the different required open loop gain. 

Anyhow, your design actively switches between the one, or the other neon, which is already or usually done by this capacitor (of about 4n7) between both neons on this neon assembly PCB A3.

This capacitor is not needed any more, and in addition complicates and threats your solution.
I suppose that it may prohibit the neons to fire correctly, i.e. at the correct phase given by your circuit, due to the bias it creates.
Maybe due to ageing of the neons (e.g. change of their striking voltage), they may only fire every other cycle, i.e at half the rate.

So you should remove that capacitor and check, if it now works correctly.

Frank
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Kleinstein on March 25, 2020, 05:32:26 pm
I don't think the capacitor should be a problem with the mod. The additional switching forces the neons to alternate, but the capacitor still works as before: it prefers triggering the other neon and gives some additional current for the initial phase. That additional current can make the LDRs a little faster to turn on.
So I don't think the mod is causing problems.

As far as I understand the plan. with the 1 V range the 90 M resistor is switched in (so there could be a problem with this resistor). The other think happening is using the trimmer R44 for gain adjustment.  Especially checking R44 may be a good idea.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: gridleak on March 25, 2020, 06:14:01 pm
The 90M resistor was the first thing I thought of. I cleaned it with IP. Didn't help. As far as A4R44 trimmer, I don't think it's a problem, or for that matter anything on A4, because I swapped in another A4 I have, with the same results. The phase reversal of closed loop gain, starting at the 3v range I reported, is per spec for the instrument (see page 4-1/Table 4-1).  That custom 90M resistor seems to be spot welded onto the instrument case.  If it's bad, then my HP419A is a total loss. I may have to bite the bullet and try to restore my other HP419A which I purchased for parts.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Kleinstein on March 25, 2020, 07:13:11 pm
The 90 M resistor is not that critical / special. The 419 is a nice meter, but not very high accuracy. So the demands on the 90 M resistor are not that high and one could get / make a new one.

With no easy fix it would be checking the circuit step by step: The first point is usually the supplies. The critical / exotic LDRs seem to work OK in the low ranges and the switches are usually relative robust. So chances are high the meter is repairable.

One point to check may be the 1 V zero pot. If this adjustment is horribly off, the amplifier may be have odd.

Another possible fault type could be oscillation of the amplifier for some odd reason. Still odd to have this with 2 versions of the A4 board. So the fault would likely be with the other parts, like the switch units or a lose cable. Still checking for oscillation would be good idea.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: gridleak on March 29, 2020, 03:01:28 am
Yes..very strange oscillation of 2vpp at 9hz occurring at at 3-- cardinal ranges (e.g. 3mv,30mv...30v,300v). All ranges have a 20mvpp 50hz oscillation but I think that may be a normal artifact of the demodulation. Unit was on battery only. If this is due to cracked or disconnected resistor on the attenuator switch, I think restoring my other 419 will be less labor. Right?
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Kleinstein on March 29, 2020, 09:28:54 am
A cracked resistor would be a rather simple to fix part. As there is not need for high accuracy there just some 9 x 10 M in series , or 4 x 22 M  could work. It may not look pretty but should be OK. The point just is if this is really the cause.

A 9 Hz oscillation is rather slow. I see two possible sources here: one would be a classic oscillation in a feed back system going unstable because the frequency compensation does not work right somehow.

The other possibility would be a beet frequency (or inter-modulation) from the chopper frequency and mains hum. Even with battery power, that is mains frequency hum around in the form of an AC magnetic and electric field unless one is in the wild really far off the grid (e.g. km range). A root cause for such trouble could be a fault in the shielding (e.g. a broken link or isolation). Another point could be the AC amplifier or demodulator getting close to saturation, thus leaving the linear range. This can reduce the tolerance to mains hum.

So far the symptoms do not look that bad and fixing this unit looks very doable.
One could check the other 419 to see how bad that one is. Only than one can decide. Chances are one could fix both. Having one unit fixed can help with the 2 nd one as one has a good one to compare the signals at intermediate levels.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: gridleak on March 29, 2020, 08:53:20 pm
The attenuator switch as a problem is now less likely. See S1A(F) (page 6.3/6.4). One can see the S1A(F) contacts resistor string S1R13 thru S1R18. As S1A(F) proceeds from 3uv to 1v, the range switch removes one resistor at a time ending at S1R17. At this point S1R18 resistor is the only resistor left in the DC feedback loop. Also, up to this point there has always been resistance in the path for DC feedback. Starting with 3mv range, the R18 resistor is removed, so the feedback loop is now a short, and instrument oscillates. Move the attenuator switch to 10mv. Now there is NO feedback on the SIA(F) portion of the switch, and no oscillation. All 3mv,30mv...30v,300v positions are shorted to feedback, and all 1mv,10mv...1000v are no feedback. Thus, with feedback closed loop shorted from S1A(F), the instrument oscillates. No feedback, no oscillation. I still don't know why this is happening, but now I see the condition that causes it.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Kleinstein on March 30, 2020, 08:45:43 am
With the difference between the 1 , 10 , 100 and 3 , 30, 300 mV ranges it in deed makes sense to looks at the switch S1A.
There still is feedback in the 1,10,100 mV ranges: these have S1R18 in series, the 3,30,300 range don't have S1R18 in series.
The 3, 30 , 300 mV ranges have the maximum feedback and this may be too much for stability.

For me the suspects would be the capacitors in the feedback loop, C13, C14 and maybe C15,C16,C17 (used to dampen the meter movement). R38 could also be a candidate to check: those high value carbon resistors tend to drift up.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: gridleak on April 01, 2020, 05:31:57 pm
You did request that I check power. This I didn't do since the neons were lit, and their waveforms were perfect.  Instead, I began disassembling the unit to get to the components you indicated could be problematic. In the process of my disassembly, I checked continuity to make sure I hadn't pulled apart connections. To my horror, I discovered that the battery pack neutral was not bonded to the instrument case! How could the instrument perform well for years with this defect?! Never mind. Bonding the battery neutral to case stopped the oscillations.

I got good advise, and I ignored it.
Shame on me.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: gridleak on April 03, 2020, 12:03:12 am
Success was short lived. In about 5 minutes the chopper warmed up and then the 3x range oscillations returned.
Again swapped A4 boards. No luck. Regardless of the perfect chopper waveform with the neons in use, I decided to change them. This time I used the GT-NE6H1925T neons which seem to be gold standard for use with 419. That didn't help. 3x oscillations continue. It is always possible that a pattern failure is affecting both A4 boards. In time, I will recap one of them to see if that helps. There is a polarized capacitor on the modulator side of the chopper assembly that couples into the ac amplifier. No value given. I don't see any capacitor back from the green lead that goes to pin 1 of A4 outside the chopper. Perhaps this is an rf style feedthrough cap. I'll take apart another chopper assembly and try to find it. I think it is a prime suspect.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Kleinstein on April 03, 2020, 08:37:39 am
In the schematics I have (scan from the web), there is C4 = 100 nF between the modulator and AC amplifier (part of the modulator assembly). At the output of the AC amplifier there is C9 = 1 µF.

I really don't think the trouble is with the neons.  The oscillation at the higher range with lot's of feedback would happen if the chopper and AC amplifier would have too much gain. A well working chopper should have high gain and a poor version would be lower gain, with very little room to get too much gain from poor neons.

If at all there could be more or less intensity effecting the output impedance of the demodulator, and this would effect the filter around C14+R26. R26 has a special mark - it is selected to get the right response time and thus effects the loop compensation . It may be a part to adjust with an aging demodulator.

When replacing C13 (4 µF) I would more go for a slightly larger value (4.7 µF) instead of less (e.g. 3.3 µF) - this could improve loop stability a little, though slightly slow down the response.

Aging of electrolytic caps could effect more than one A4 board. Still odd to see a similar error with 2 boards.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Dr. Frank on April 03, 2020, 03:25:01 pm
The attenuator switch as a problem is now less likely. See S1A(F) (page 6.3/6.4). One can see the S1A(F) contacts resistor string S1R13 thru S1R18. As S1A(F) proceeds from 3uv to 1v, the range switch removes one resistor at a time ending at S1R17. At this point S1R18 resistor is the only resistor left in the DC feedback loop. Also, up to this point there has always been resistance in the path for DC feedback. Starting with 3mv range, the R18 resistor is removed, so the feedback loop is now a short, and instrument oscillates. Move the attenuator switch to 10mv. Now there is NO feedback on the SIA(F) portion of the switch, and no oscillation. All 3mv,30mv...30v,300v positions are shorted to feedback, and all 1mv,10mv...1000v are no feedback. Thus, with feedback closed loop shorted from S1A(F), the instrument oscillates. No feedback, no oscillation. I still don't know why this is happening, but now I see the condition that causes it.

There is a possible hint in the manual, Theory of Operation, page 4-2, paragraph 4-24, about the gain of the AC amplifier.
"AC feedback from the emitter of Q3 to the emitter of Q1 is used to vary the the gain of Q1 through Q3...in the ranges 3µ - 1mV, resistor R7 is shorted out, decreasing the negative feedback applied to Q1."

That means, in the upper ranges from 3mV onwards, the AC gain should be reduced as less total open loop gain is required.

If that doesn't work correctly, the overall amplifier may oscillate.
Therefore,  please check S1B(R), if that shorts R6 correctly to GND in position 1 - 6 only.
Is C2 i.O., does it still have 200µF and low ESR value? Same goes for C6, C7, which might also influence AC gain.
I assume, that you already checked all other electrolytics, which smooth the supply voltages.

Are the expected DC level around Q1..Q5 ok, and all resistor values, R1..R22?

What's still strange is that this oscillation starts after warm-up, so it may have something to do with the neons also, maybe you try anyhow what happens if you desolder C1 on A3 board.

Frank
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: gridleak on April 08, 2020, 12:49:09 am
All electrolytics pass ESR.
Switch S1B(R) does short out R6 only in the first 6 positions.
C2 286uf - ok also C6,C7 ok
Collector voltages Q1-.5v Q2-.55v Q3-3.4v Q4-3.9v Q5-.62v
Q5 is a little low compared to schematic value .75v.
I reduced A4R26 from 10k to 5k No luck.
Checked switch resistors on top of switch eg. R12 thru R20 and R21 thru R23, R25,R27 those all ok.
Checking bottom resistors of switch not practical without disassembling pushbutton platform, so I think.
I don't understand that capacitor C13 on demodulator feedback side to dc amplifier affects the stability in ac amplifier feedback after modulator which is where oscillation is taking place.
Not able to perform further investigations due to battery pack for -13.2 volts is no longer holding charge. Will have to order a new battery. This is a power tool battery pack. Hope will be back up soon.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: Kleinstein on April 08, 2020, 08:57:21 am
The DC voltage for the AC amplifier (Q1-Q5) look OK except for the collector of Q4 - probably a mistake or due to the -12 V supply allready down to some - 4.5 v.
The -0.62 V at the Q5 collector are perfectly OK, as there is AC coupling at the output.

C13 is the main part in the DC amplifiers feedback. The DC amplifier with C13 is configured as an integrator.
The  path with CR5-CR12 is there for limiting the maximum amplitude and should normally not be active.
The direct DC path through R38 is very weak:  With R37,R39,R38 and C13 the time constant is at some 160 seconds.
I would think one would not even notice if R38 is missing. It may be there only for tests with the chopper off.
For understanding the circuit is may help to look at the simpler (e.g. older A1 version).


There is too much feedback that causes oscillation. So to stop it, a larger value (e.g. 20 K) for R26 could fix it, not a smaller one.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: gridleak on April 08, 2020, 12:24:32 pm
My fault that I put dashes after each transistor. Only the Q5 is negative voltage. My copy of schematic shows +3.9 volts for Q4 which is what I measured. Since I can't do more investigations until I have a new battery, I will spend more time understanding the circuit operation. Meter is very sluggish at 3uv range. Instructions are to lower A26 to speed up. I fear that needle will never settle with A26 raised up. Still, will make the change upwards to see if oscillations stop as result.
Title: Re: "GREAT" Meter! The HP 419A DC Null Voltmeter, [RESTORATION]
Post by: ArthurDent on May 22, 2020, 05:35:58 pm
My HP 419A is basically unmodified except for the new batteries I put in some time ago. I bought four of the 5 cell AA NiCd battery packs on eBay designed for phone use and they fit in nicely and work well. Here is a Youtube link of the meter comparing two 10.00000 volt LTZ1000ACH references with the null meter set for 3uV full scale. Also added a photo of the replacement batteries in place.

https://www.youtube.com/watch?v=wTGxBJNNhWc (https://www.youtube.com/watch?v=wTGxBJNNhWc)