Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: Prev 1 2 [3] 4 Next   Go Down
  Print  
Author Topic: TL922 Grid Current  (Read 18325 times)
N3QE
Member

Posts: 5576




Ignore
« Reply #30 on: March 06, 2014, 01:49:59 PM »

Tom -
  While textbooks and your website show the metering circuits very simply in a way that most hams can grok, by the time a manufacurer introduces a multi-meter switch and then splits the function between circuit boards and wiring harnesses, the resulting manufacturers schematic is often un-grokkable.

  Incidentally in my AL-811H when the grid current meter started inexplicably pinning, I found the problem to be an short between meter movement and meter lamp! Opening up the meter and re-coiling the meter lamp wire solved that one.

Tim
« Last Edit: March 06, 2014, 01:56:39 PM by N3QE » Logged
W1QJ
Member

Posts: 2945




Ignore
« Reply #31 on: March 06, 2014, 01:55:43 PM »

I should stress, as Peter said that the series resistor has to be outside the switch loop for the meter.  He'll have to lift the wire going to the switch from the meter shunt and insert the resistor there.  As I recall the 922 uses a series reistor for Ip but not for Ig.  For Ig the shunt was calculated for the actual meter FS voltage.  The SB-22o is the same way.  Gri current direct off the shunt and has a 3600 ohm series resistor with the 1 ohm Ip shunt.  He can accurately measure the resistance with his nice DMM and can measure the FS voltage with a simple jig and measure the voltage across the meter at FS.  I use a battery and a pot remove the wires from the meter, bring the meter up to FS and measure the voltage across it.  Then I can calculate the series resitor based on the installed meter shunt.  Pretty simple.
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #32 on: March 06, 2014, 05:32:53 PM »

I should stress, as Peter said that the series resistor has to be outside the switch loop for the meter.  He'll have to lift the wire going to the switch from the meter shunt and insert the resistor there.  



My answer was correct. A series resistor at the meter to correct grid current has NO observable effect on the other two meter functions.

I really didn't want to do all this work, but here it is.  Smiley  Let's do the actual calculations and figure out the values:

The meter is 4kV and 400 mA FS, the multiplier is 3X 2.66M ohms, and the shunt is .68 ohms.

HV 4kV
 
The HV is measured through a series resistance of 7.98 megohms. The current through that path at 4 kV is .0005 amps. The meter is a 500 uA meter.

Grid Current 400 mA

The grid current is .4 amps for full scale.  A 0.68 ohm resistance would drop .4*.68= .272 volts. This means the shunt is a 272 mV shunt

Meter movement

The meter movement must be .272 / .0005 = 544 ohms. So the meter movement is a 272 mV, 544 ohm, 1/2 mA movement.

Now let's install a standard 500 uA 100 mV movement (I don't know what he used). The resistance is 100 ohms.

Actual Circuit
The meter is paralleled with 47K on HV. That means the 544 ohm meter and 47k looks like 538 ohms. 4000/7980538 = .0005012 A, with about 1% flowing through the 47K. The meter will read full scale.

Now let's assume he replaced the movement with a 100 ohm movement.

The meter and 47K will now look like 99.8 ohms. The current will be 4000/7980010 = .00050123 A of which 0.2% flows through the 47k.  The meter changed about 0.8% with this 444 ohm meter resistance change in the volt position.

Now let's look at the grid current:

The 100 ohm meter would have .05 volts full scale. The .68 ohm shunt would make it .05/.68 = .07353 A, or 73.53mA, full scale for grid current. The change from 544 ohms to 100 ohms in the meter changed the grid current FS from 400 mA down to 73.53 mA. The HV changes less than 1%, which he would never notice, while the grid scale to 543% of the original scale reading.


What I have been saying is correct. I didn't want to do all the work so I gave typical examples, but these are the Kenwood numbers.

He probably has replaced the movement with the wrong sensitivity movement, but the correct current. The way to fix this would be to add a resistor in series with the movement at the meter or inside the meter.

Once again here is the rule on meters:

1.) A meter movement used in a normal high voltage meter is current critical. Resistance does not matter much.

2.) A meter movement used on a shunt to read much higher current is voltage critical. Resistance and current equally matter.

3.) If someone changes a meter and they report the voltage seems to read correct but the current is off, they almost certainly used the wrong resistance movement but the correct current. The FIX is to make the movement have the correct resistance, so the resistance needs to be at the meter.

73 Tom
Logged
W1QJ
Member

Posts: 2945




Ignore
« Reply #33 on: March 07, 2014, 04:01:31 AM »

GM Tom, yes, I stand corrected, as you put it, it is obvious it can be done like that with no ill effect.  You didn't have to go through all the math but it sure is a great tutorial for those who don't know how to calculate these things.  If someone learned something from it, then it was worth it.  Lou
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #34 on: March 07, 2014, 04:23:09 AM »

GM Tom, yes, I stand corrected, as you put it, it is obvious it can be done like that with no ill effect.  You didn't have to go through all the math but it sure is a great tutorial for those who don't know how to calculate these things.  If someone learned something from it, then it was worth it.  Lou

Thanks, Lou.

The motivations behind this were not to argue, but to prevent another old hacked up amplifier from appearing for someone to figure out later, and so anyone else who repairs or replaces a meter understands how it works.

The replacement meter, if it is scaled as a voltage meter with a fairly high voltage, has to have the same current rating.
The replacement meter, if used with a shunt as a current meter for a fairly high current, has to have the same voltage sensitivity.

If it is a multimeter, it has to have both. So when he puts the resistor on the meter, it is back like the original circuit. Later on someone else doesn't have to wonder why the heck the resistor way off somewhere else was needed.
Logged
W1QJ
Member

Posts: 2945




Ignore
« Reply #35 on: March 07, 2014, 05:37:36 AM »

GM Tom, yes, I stand corrected, as you put it, it is obvious it can be done like that with no ill effect.  You didn't have to go through all the math but it sure is a great tutorial for those who don't know how to calculate these things.  If someone learned something from it, then it was worth it.  Lou

Thanks, Lou.

The motivations behind this were not to argue, but to prevent another old hacked up amplifier from appearing for someone to figure out later, and so anyone else who repairs or replaces a meter understands how it works.

The replacement meter, if it is scaled as a voltage meter with a fairly high voltage, has to have the same current rating.
The replacement meter, if used with a shunt as a current meter for a fairly high current, has to have the same voltage sensitivity.

If it is a multimeter, it has to have both. So when he puts the resistor on the meter, it is back like the original circuit. Later on someone else doesn't have to wonder why the heck the resistor way off somewhere else was needed.

Tom, if you read my earlier post about how N6WK and I cloned the Ip meter for the 922 You'll notice I said that we supplied a resistor to replace the series resistor that is already in there.  I thought about trying to keep the existing resistor and work wtih just the meter but as I recall he series resistor we needed was a lesser value than the series resistor already there.  So we were forced to change the existing resistor.  Had  it required a higher resistance we could have added it inside the meter and there would be no circuit change if that meter ever got replaced with a stock meter. We didn't have the pleasure of finding a meter that was the right size with a red needle that would  look almost exactly alike.  I must say it was a pretty good match. 
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #36 on: March 07, 2014, 06:47:43 AM »

I certainly don't think a single resistor added on is an issue.

The real problems are if the meter resistance is off the voltmeter functions are off very slightly, the current is way far off, and pots just look cobbled up.

More than any of that, I hope we want to learn how things work. I view amateur radio and forums as educational places, and the CB band and forums as places to not learn.  Smiley
Logged
K6JEY
Member

Posts: 11




Ignore
« Reply #37 on: March 09, 2014, 06:14:53 PM »

To Tom et al,
     Here are the values measured on the amp.
       Movement resistance is about 177 ohms (too high by maybe 1 ohm )
       Full scale voltage to read 400ma on the meter- 88.7mv
       Full scale current to read 400ma on the meter 0.50ma
 
      The regular shunt is measured at .680ohms (+-. 003 ohms)
   
      Please let me know the additional resistance to add in series.
 
      Doug K6JEY
Logged
K6JEY
Member

Posts: 11




Ignore
« Reply #38 on: March 09, 2014, 06:28:13 PM »

Hi Tom, Following your excellent instructions, I come up with a needed resistor of 368 ohms.  Is that correct?
     Again your time, experience and brains are most helpful. You are right, with just a resistor added, it won't look too cobbled.
     The place I picked up the measurement was at the IG post on the resistor board. the thin brown wire goes directly to the meter switch in the IG position. There was probably a slight bit of resistance added to the meter resistance as a result. I took off an ohm to compensate the measurement.
     Was I close?   Doug
Logged
W1QJ
Member

Posts: 2945




Ignore
« Reply #39 on: March 09, 2014, 08:31:46 PM »

Yup, I got 366 ohms so thereabouts would do it
Logged
W1QJ
Member

Posts: 2945




Ignore
« Reply #40 on: March 09, 2014, 10:45:50 PM »

BTW  Doug, there probably is already a resistor inside that meter.  You might want to look at it, see the resistance and add that amount to what we calculated and just put one resistor in there.
Logged
K6JEY
Member

Posts: 11




Ignore
« Reply #41 on: March 09, 2014, 11:22:06 PM »

Hi, Here is what I came up with so far. I put in a 360ohm resistor in line and it totally clamped the meter. I put in a small trim pot at 500ohms. With just a little effort I found the right resistance and the meter is working. The problem is that the meter switch is hidden waaaaay down between the chassis. The meter itself is also buried between the front panel and the chassis. The final run will be to put in the current meter on the main lead and adjust the pot to match the reading. Bottom line is that the resistor was too hi in value. (it was accurate, just wrong value)  However, thanks to Tom, weeks of trying to sort out the problem has resulted in a very minimalist fix and gotten a pretty badly traumatized amp back on the air.
      Doug K6JEY
Logged
W1QJ
Member

Posts: 2945




Ignore
« Reply #42 on: March 10, 2014, 02:59:16 AM »

Doug, you are quite right, it is a chore to change a meter in that amp.
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #43 on: March 10, 2014, 07:01:29 AM »

Hey guys, remember I guessed at the meter resistance in Doug's amplifier because I have no idea what meter movement he used. I just picked a standard movement. He needs 544 ohms of total resistance in the meter.

Who rebuilt the meter, Doug, and did you measure the resistance? You can measure the resistance with the meter in circuit by selecting the HV scale. In the HV position it is paralleled by 47k, which is nearly meaningless.

Doug....this isn't the first time a meter has been repaired in a way that makes some readings off. I just wanted to explain why these funny readings happen when a meter is swapped, and how to fix it or prevent it.
Logged
W1QJ
Member

Posts: 2945




Ignore
« Reply #44 on: March 11, 2014, 04:42:30 AM »

Hi, Here is what I came up with so far. I put in a 360ohm resistor in line and it totally clamped the meter. I put in a small trim pot at 500ohms. With just a little effort I found the right resistance and the meter is working.

Doug, I am pretty sure I know which clone meter you have.  The reason why the calculated resistance did not work is because if my hunch is right the clone meter is 720ua FS.  That would require less resistance than calculated, hence the meter was clamped.  I pulled one of those meters and tested it yesterday and realized it was not 500ua.  That does not explain why the HV is reading correctly though.  Unless someone changed out the series resistors.  If the clone meter is 720ua like I measured then the series resistance would be 5.55megs  When Gordon and I figured out the system for the plate meter that is why we had to supply a separate resistor along with the meter.  The resistor we supplied replaced R7 because R7 was already to high in value.  Had it worked out to be lower we could have added the difference inside the meter.  The wisdom is this, If you blow a multimeter out you should remove both meters from the amp, switch the scale on the plate meter for the multimeter and use the clone meter for plate current since it only has one function.  If the demand is there I will once again make the clone Ip  meter available.
Logged
Pages: Prev 1 2 [3] 4 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!