a timtron turbo mod question

[NR5P]

Well I rerouted the wire the way I should have to begin with.  Shortest run possible and it has fixed the oscillation problem.  but now don't have any modulation when I increase mic gain.  It's easier to figure out atleast, I replaced the 12ax7 and the 12by7 in the audio chain as a quick check and it's something else.  I sure have learned alot working on this thing.  If only it weighed 10lbs instead of 100

[NR5P]

just an update with a dmm I checked across the grids with the el34s unplugged and getting about 2 volts a/c while talking in mic on the meter.  I know it's not the most accurate way but tells me there is audio getting there.  It doesn't surprise me as I didn't mess with that part of radio.  Gonna do some more rechecking of what I wired.  I really wanted to get on the air tonight with it!

[NR5P]

I have double and triple and quadrouple checked using an ohm meter and looking at the wiring in the radio to make sure the wiring I did was correct and it appears to be.  I guess I need to check again using the 120v method that I have the correct plate connected to 500ohm tap, but I was careful in checking that and was happy that it was the tube plate that was easy to get to when I checked so it wasn't a problem getting the soldering iron in there.  tubes are idling correctly now and appears that modulation is getting to them.  I'm beginning to run out of ideas here.  I checked the screen and plate voltages on the el34s also and that is fine.  and grid dc is also good.

[WD5JKO]

I have double and triple and quadrouple checked using an ohm meter and looking at the wiring in the radio to make sure the wiring I did was correct and it appears to be.  I guess I need to check again using the 120v method that I have the correct plate connected to 500ohm tap, but I was careful in checking that and was happy that it was the tube plate that was easy to get to when I checked so it wasn't a problem getting the soldering iron in there.  tubes are idling correctly now and appears that modulation is getting to them.  I'm beginning to run out of ideas here.  I checked the screen and plate voltages on the el34s also and that is fine.  and grid dc is also good.

Does the modulator cathode current kick up when you talk? If not do the cathodes find ground through the metering circuit and 0.2 ohm resistor? Or maybe it is modulating fine, but the meter does not move..If all that checks out, all I can say is reconfigure back the the last known working condition, i.e. remove the turbo mods..

I'm beginning to think there is more than a typo or two in that mod. as stated.

I live about an hour away from you, and I do have Monday off....


Jim
WD5JKO

[Opcom]

Roger OM my Modulator tubes here are a 50C5 in the neg side and a 811A in the positive side.   

I want to learn your methods to make that work! And here I just wanted to make a little more current available to the side that was short on magnetization. I obviously have no ambition!

I'll try removing the driver next.  I've only done the w3scc mod and turbo mod.  I also have not added any NF.  I did check the phasing both ways I connected 120v across the modulation tubes(primary of transformer) and jumpered the center tap to each plate.  One I got 98volts on(thats the one I permanently connected) the other read much lower I can't remember now what it was. 

If this is on that same TX-1 transformer, then by the specs shown in the schematic, putting 120VAC the whole pri. should make the whole sec. do 62.4V, and from the 500 Ohm tap to the hot side ought to be 37V, and to the cold side 25.6V.

So, 1/2 the primary being 60VAC, and adding the 37V coming from the 59% connection (500 Ohm to 'hot'), there is the 97v, same as your 98V. Perfect!

I have double and triple and quadrouple checked using an ohm meter and looking at the wiring in the radio to make sure the wiring I did was correct and it appears to be.  I guess I need to check again using the 120v method that I have the correct plate connected to 500ohm tap, but I was careful in checking that and was happy that it was the tube plate that was easy to get to when I checked so it wasn't a problem getting the soldering iron in there.  tubes are idling correctly now and appears that modulation is getting to them.  I'm beginning to run out of ideas here.  I checked the screen and plate voltages on the el34s also and that is fine.  and grid dc is also good.

The post you made previously shows you had the larger amount of boost. If the tubes biasing and current is now behaving right, then the oscillation if that was the main issue seems to have been dealt with.

If all this is fine, are you not getting sufficient modulation now? If that is the case, the best thing to do might be to use a scope and look at the modulator plate (where the boost winding is connected) and note its most negative and most positive voltage when beeping a sine wave in there. Do the same at the hot end of the mod transformer secondary too.

With a B+ of 750V, you need +1500V peak to be delivered to the RF stage for 100% positive modulation. That is either stock or turbo'd, it does not matter. With the added 59% of your boost secondary there, it should add 460V to that and give 1960V peak.

To check it safely with a scope, just tack on a resistive voltage divider made of five  1M resistors, and tap the scope off the lowest one, so when you read 200V safely on the scope you know the circuit voltage is 1000V.

If RF is an issue to the measurements:
A small capacitor to GND cross that first resistor will help get rid of any RF messing up the measurement. Maybe 100pF or 0.001uF? Not too large though or it will load the voltage divider and throw off the measurement.
Or
Remove the final and put in a resistor equal to 750V/175mA (as Tim stated hopefully that's OK) 4300 Ohms @ (at least) 132W.

Just some ideas, can't swear by the outcome. More than once thougn a second pair of eyes on something was just what was needed when I overlooked something and could not figure it out.

[Gito]

Hi


I don't have the Heatkit apache schematic,but I have  the heatkit DX 100 (?)looking at the mod circuit ,the modulator tube is a push pull circuit which is drive by an inter stage transformer

Case 1 : with  a certain connection .the  modulator transformer has a current flow in phase with the current flow in the inter stage transformer.(output voltage in phase with input voltage}

  Case 2:when swapping the plate connection the current flow in the Mod.Trafo is 180 out of  phase with current flow in the Inter stage trafo.(output voltage is out of phase with the input voltage

In these two cases .the out put voltage in connection case 1: is 180 degree  againts the output voltage in case 2.

In case 1: an audio signal(x) is put into the audio circuit,audio -interstage transformer- modulator tube -transmitter---RF got into the audio circuit--rectified  --rectified audio + the original audio(x) ----input audio.

Since this audio signal are in phase  (rectified audio + original audio) than the audio amplifier began to oscillate

Gito.N

[NR5P]

I'm going to reread over these last few posts and maybe do some checking tomorrow.  To answer wd5jko though it definately isn't modulating as I have the power meter inline also.  I appreciate your offer I've got some other things to do around here monday and hate to use up your day off. 

When I hook up the resistor chain to check with the scope I put them to ground correct and ofcourse it would be the one closest to ground to tap off of. 

But yes the problem now is I'm not getting any current at all when I talk into the microphone.  Seems to be idling fine at 50ma.

[NR5P]

Well a little while ago I hooked up mod transformer the way it was(easy enough to do) to see if my problem is related to the mod. I still got zero modulation registering on the meter.  So I stuck in the extra el34 I have (only other el34 I have) and removed the mullards.  Then I tried again and I got slight modulation from the one tube when i turned up the gain I only had it idling at 25ma because of one tube.  Maybe there is damage to the tubes from the oscillation problems I was having. 

I did notice something I have not noticed before.  when I place the meter selector in rf final plate current or modulator plate current(even if the modulators are removed) after about 5 seconds the meters slowly go below zero and start lightly bouncing off below the zero.  Maybe this has been doing this all along and my bias I am adjusting is much lower that what I'm adjusting.  I can double check this easy enough with a seperate meter but for now need to find out whats going on with the negative charge building up.

I'll be in the radio room throwing stuff and kicking radios

[NR5P]

Ok I forgot and didn't have the ground attached to the chassis.  I think I am narrowing down the problem and things look better.  When I crack the mic gain all the way open I'm getting slight modulation on the current meter.  Must be something in the preamp sections.  Either with the 12au7 or 12ax7.  I've tried replacing the tubes earlier and no dice so gonna go through it maybe tomorrow and see if I can find where the problem is.

[NR5P]

Great!!!  I found the problem the "spring" that is around the audio connection that is a shield was touching one of the heater voltage pins on the 12by7a.  I guess it wasn't causing enough of a short to blow a fuse but was drawing down on the transformer.  It supplies voltage to all of the audio tube heaters.  I hope I haven't shortened the life of the transformer.  These radios can sure take  a beating.  Im going to make the connection again now for the timtron mod and I will let you know results.  I guess I'll be on am this weekend.

[Opcom]

Good! I hope that's all it was. I'll be happy for you anyway if it's got the turbo thing going.

[Gito]

Hi


This Turbo modulation troubled my mind ,with the original connection ,the voltage ratio of the modulation transformer is 627 : 1200  (1 : 1.9 },the impedance ratio is 3K to 11K.
 with the turbo connection the voltage ratio becomes (600 + 370) : 1200 = 970 :1200 = 1.237 ,the impedance ratio is 1 : 1.530.

Since the modulation impedance of the transmitter is 3 K . ( 750 : .25)
Than the reflected  impedance 1.530 X 3K = 4.593 K

The EL 34 operated  as modulator in the transmitter condition (original connection} ,has a 11K  impedance,the right load.

But with the Turbo Modulation  The EL34 looks at a 4.593 K load ?(reflected impedance) not the right load.

So this puzzled me? Do I missed something

Gito.N

[K1DEU]

 Not much Time for me. But I look mostly at magnetics and coils of wire with and without cores that may reduce the physical size of a coil. Unfortunately there are so many different cores that it is very difficult to find some material that collapses and builds up a near field with out adding distortion. Sometimes we want the field to be misshapen on purpose. Tubes and transistors are mostly current switches. And can be harmed by requiring them to exceed their heat dissipation or voltage max ratings. 73  John

So you want to discuss impeadence ?  Well you have Walt W2DU to guide you.

[Opcom]

Hi


This Turbo modulation troubled my mind ,with the original connection ,the voltage ratio of the modulation transformer is 627 : 1200  (1 : 1.9 },the impedance ratio is 3K to 11K.
 with the turbo connection the voltage ratio becomes (600 + 370) : 1200 = 970 :1200 = 1.237 ,the impedance ratio is 1 : 1.530.

Since the modulation impedance of the transmitter is 3 K . ( 750 : .25)
Than the reflected  impedance 1.530 X 3K = 4.593 K

The EL 34 operated  as modulator in the transmitter condition (original connection} ,has a 11K  impedance,the right load.

But with the Turbo Modulation  The EL34 looks at a 4.593 K load ?(reflected impedance) not the right load.

So this puzzled me? Do I missed something

Gito.N

Hello Gito,

I apologize for these numbers because I had them wrong in the drawing due to not knowing the TX-1 B+ was 1500V. I replaced that drawing in the previous post, and here it is for convenience. I am very sure it is correct assuming "perfect" tubes (in reality the plate swing does not go to 0 on the modulator tubes because of their G2 voltage). But it is the turns ratios than matter.

I agree the original connection is 627:1200, or 783:1500. (1.044 or 1.045 for secondary to half primary)

I have to think about the impedance ratio in the turbo hookup but if the swing to the RF stage is the sum of:

0 to 1500 of the modulator tube,
plus
-462V to +462V,

then it is:
0 plus (-462V) to 1500V plus 462V
which is -462V to +1962V.

(diodes are assumed to prevent negative modulation beyong 100% from the previous article)

If the total winding of the primary is 11K, and the secondary is 3K (by the turns ratio or the specs) and only 59% of the 3K winding is in use then the secondary portion in use is 2500 Ohms.

There are issues with saying this because the windings do not stack impedances, only voltages, so it is easier to look at it from a voltage standpoint.

If the total primary has 1500V peak on it and the partial secondary has 462V peak on it
and
If the primary is 11K CT then, the signal ground is the CT, so that makes the primary impedance 2.75K to ground.

1962V / 1500V = 1.308 voltage ratio
1.308 * 1.308 = 1.711 impedance ratio
2.75K * 1.711 = 4.7025 K

I think our numbers agree closely enough (4.593K to 4.703K) about the impedance the RF stage should present, IF we want to see 11K CT at the modulator tubes.

It was said that the plate current had to be reduced to 175mA to avoid saturating the transformer due to DC imbalance.

The RF stage then becomes 750 / 0.175 = 4286 Ohms instead of 3K.


mine:
11K / 4.7025K = 2.339
yours:
11K / 4.593K = 2.3949


750V @ 175mA = 4.286K

mine:
4.286K * 2.339 = 10.0249K plate to plate
yours:
4.286K * 2.3949 = 10.2645K plate to plate

So the EL34's will work harder, but a 10K or 10.3K plate to plate load should be OK.The load would be too light with the RF stage at 750V/175mA if not for the added winding. With the given winding it is a bit heavy by 8-9%, so let the tubes show some color maybe?

This brings up another point related to the reduced input @ carrier, unless I misunderstand it. The reduced input was to be arrived at by drive or loading changes.

the input will be 70% of the normal input.

The theoretical PEP available at 130% positive modulation would therefore be not 6.8x stock but 4.76x stock.

I think that is 118% positive modulation but not sure. There may be some tuning to help this, but the unbalanced DC thing is going to come back, and the plate to plate load on the modulators is going to decrease a lot if the stage is tuned back up. Tim said 125% mod worked, maybe there is a middle ground where it works best.

If there were no issue with core magnetization, some other issues begin to arise that have not yet been discussed:
-total current capacity of the power supply
-current capacity of the modulation transformer primary
-ability of the modulator tubes

So, the results will be interesting to see how well this works.

[Gito]

Hi


The way I see it,the voltage developed across The modified circuit  for the transmitter is the voltage  of the CT of winding of the modulator tube  + the voltage the primary winding minus the 500 tap winding.

750 + 462 volt = 1212 volt since the reference volt(B+) is 750 volt than the voltage developed across the transmitter is 750 + 1212 =1962 v  positive peak modulation
 and the the negative peak modulation is 750 - 1212 = -462 v.

since the AC voltage from the CT (B+)  to the plate of The RF Pa is always plus and minus 1212 v.

Than the turn ratio is 1500 : 1212 =1.237 or the impedance ratio is1.5317.

If we load the transmitter with 250 ma ,than the impedance is around 3 K ( modulating Impedance} Than the transmitter looks into a 1.5317 X 3 K= 4.5951 K.

Remember the the winding from the CT to the Rf Pa is a series Winding,and in phase

and CT is the reference point, so a positive swing across the modulator winding will cause a positive swing across the CT to the RF pa.windings.
A negative swing across the modulator winding will cause a negative swing across CT to the RF PA windings.

But of course I can be wrong

Gito.n

I modify and attach a picture ,that maybe more explaining ,sorry for my pictures ,I can't make good pitures

[Opcom]

(perfect parts assumed and screen voltage issues ignored) I agree except that I believe the peak to peak voltage across the modulation transformer primary is 3000V. I believe this because I believe that the plate of each modulator reaches a 1500V peak voltage and each modulator plate swings from zero to 1500V. That is why I said 1500 + 462V.

I also add that it may be me who is missing something, so if so please point it out. I'm no expert on this stuff.


You can make good pictures on even the most backwards PC! (8086, CGA, mouse, pc-painbrush) --> The pictures were made with the Microsoft "paint" program to open the file with paint, save it, then save as PNG for small data size. The symbols come from the attached file. The file is ancient from a forgotten web page but has grown a little over two decades. (I sometimes wonder how others have grown their copies or forgotten them.) Just copy the components one at a time from the file and paste each to the new bitmap file (your good drawing), and use the lines to connect and draw the schematic. It is low tech but that is what was done, so looks OK.

[NR5P]

Finally got it checked on the scope.  It definately helped out the peaks.  I'm getting right at 100% before it starts to clip.  I was getting about 80% before when this happened.  I do start to get clipping on the negative peaks at about 80%.  I was getting clipping there before the mod also.  People tell me I sound good just don't have the lows.  I've got 3 megaohms across the grid that the d104 looks into.  I've heard some people say the d104s do best into 10 megohms atleast.  Don't know if this would make a noticable difference or not.  Also may be to much across the grid and make the tube unstable. 

[Opcom]

I don't understand why you are getting only 80% on the negative side. Are you using the 3-diode limiter?

[NR5P]

sorry i worded that wrong.  When positive peaks are at about 80-90% some of my negative peaks are clipping. 

[NR5P]

They are clipping at 0 volts not 80% just to clarify again.  No I'm not using a negative diode limiter.  Radio is stock except for the w3scc audio mods and the turbo mod.

[WD5JKO]

They are clipping at 0 volts not 80% just to clarify again.  No I'm not using a negative diode limiter.  Radio is stock except for the w3scc audio mods and the turbo mod.

Ok Nathan,

  Soo, I wonder what happens if you swap the wires going to the D104 mic element? It would be neat if you then could peak 120% positive when hitting 100% negative...

  Also try swapping modulator tubes in case one has more emission than the other..

Have a tone generator? If not here is a free PC based one, SigGenFreeWare:

http://www.dr-jordan-design.de/Downloads.htm

Jim
WD5JKO

[NR5P]

Thanks no I don't have a tone generator except my voice I will give it a try.  The d104 isn't running into a balanced amplifier one end is grounded.  Would swapping the element wires still make some kind of a difference?

[WD5JKO]

Thanks no I don't have a tone generator except my voice I will give it a try.  The d104 isn't running into a balanced amplifier one end is grounded.  Would swapping the element wires still make some kind of a difference?

Nathan,

  The DR. Jordan tone generator will work with your PC. You just need to scale the voltage to your input with a voltage divider. If you get hum, use a cheap RS 600 ohm to 600 ohm audio transformer for isolation, or use a laptop running on batteries.

  Yes swapping the wires on the D104 will work with an unbalanced feed since the D104 element outputs are not referenced to ground until you hook wires to it.

Jim
WD5JKO

[NR5P]

I think I will have much better lows now.  I did try swapping wires on the d104 element and didn't seem to help.  Neither did swapping the tubes.  I did swap the d104 head to a different one with a different element and wow when I speak lower they are up there with the highs now.  Before if I said a 2 and held it it would not be nearly as loud as a one.  Seems to be greatly different frequency response.  I haven't used the tone generator yet I will need to take the computer in there.  There definately seems to be something strange with the negative peaks.  When I hold a tone with my voice or speak normally I'm seeing bottoming out on negative peaks and 100% mod is maximum even when swapping tubes or element wires.  Maybe it would take a constant solid tone to get an accurate check.

[WD5JKO]

I think I will have much better lows now.  I did try swapping wires on the d104 element and didn't seem to help.  Neither did swapping the tubes.  I did swap the d104 head to a different one with a different element and wow when I speak lower they are up there with the highs now.  Before if I said a 2 and held it it would not be nearly as loud as a one.  Seems to be greatly different frequency response.  I haven't used the tone generator yet I will need to take the computer in there.  There definately seems to be something strange with the negative peaks.  When I hold a tone with my voice or speak normally I'm seeing bottoming out on negative peaks and 100% mod is maximum even when swapping tubes or element wires.  Maybe it would take a constant solid tone to get an accurate check.

   Yea, many of the old D104 mic cartridges are crapping out, and Astatic does not make them any more. After they get old, expect differences in them. Your transmitter issue limiting modulation peaks is interesting. Maybe try boosting grid drive to the RF final to see if that helps, and/or load the final amplifier lighter.

PS, for just sine wave drive, maybe extend the wires to remote a PC speaker over to your rig, and set speaker next to the Mic element. Somewhat crude I know..

Jim
WD5JKO