The AM Forum
March 28, 2024, 01:41:35 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1] 2   Go Down
  Print  
Author Topic: This should be interesting  (Read 15796 times)
0 Members and 1 Guest are viewing this topic.
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« on: October 28, 2021, 04:00:06 PM »






After a decade or so, I'm finally got around to working on my "BigRig". (The name "Fabio" has been taken so I'll come up  with something else.) Following a consultation with an unnamed op from near Mirror Lake, I'm giving up on the Freed transformer for now; I'll be using a Peter Dahl transformer for the Modulator and RF deck power supply.

In the recent past, I've cooked the Mod transformer overnight, and the Mod Deck w/o HV for a few hours. See the pic, how nice the 810s and the 6CA7s glow in the dark.
The Peter Dahl is rated for 2135 V , .5A CCS. It's a split primary, and I got ~ 1 KV, no load. Keeping the transformer at 1 KV, I have an unknown choke, and a 23uF oil can that I propose to fire everything up and see what 'appens. The audio load consists of (3) 2K ohm, 200 Watt resistors. I'll feed the HV to the Pri and 2nd of the mod transformer so the won't feel shunned and decide to crap out. I'll scope it an see what thinks look like at 1 KV.

So, do anyone have any input/comments/ridicule r/e this thing of mine?

klc
Logged

What? Me worry?
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #1 on: October 28, 2021, 05:20:20 PM »

eL Mod deck at night.


* Mod deck.jpg (3443.67 KB, 3024x4032 - viewed 471 times.)
Logged

What? Me worry?
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #2 on: October 28, 2021, 05:23:15 PM »


Mo power.


* The power lineup.jpg (4873.1 KB, 4032x3024 - viewed 446 times.)
Logged

What? Me worry?
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #3 on: October 29, 2021, 09:58:22 AM »

What is the overall approach? Got a schematic?

What is the history? Did it get that patina from your decade of inactivity or is it a chassis from another brother...

John
Logged
W7TFO
WTF-OVER in 7 land Dennis
Contributing
Member
*
Offline Offline

Posts: 2521


IN A TRIODE NO ONE CAN HEAR YOUR SCREEN


WWW
« Reply #4 on: October 29, 2021, 11:15:33 AM »

Kev, your problem is not enuf old paint on the workbench. Sad

73DG
Logged

Just pacing the Farady cage...
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #5 on: October 29, 2021, 11:54:32 AM »

Yes, a brotha from a different muther.

The plan is to throw together what I have and see what sticks.. I started this around 2006? For Now, I'd like to put this on the air on 160m and 75m, hopefully before The Rally in February.
https://amfone.net/Amforum/index.php?action=gallery;su=user;cat=14;u=714

I picked up a 'kit of parts'. i.e. partially disassembled home brew transmitter, along with some misc. parts and paperwork that sometimes matches what I have. So, originally, an 810 modulator, BC 610 mod transformer, some power iron. For the RF side, two 813s in  Push Pull,  link coupled. The circuits are apparently from The Radio Handbook, 14th ed.

The mod deck  as found;  mic in, 12AT7 then a 12AU7 that tickles the grids of two 6CA7, which then feed the 810s.

The Plan.   Power supply design for the mod deck and RF Deck. Then then the RF deck.

"JS"  a full wave rectifier, with the transformer pri. series wired to provide ~ 1 KV, add the unknown choke and use a 10 uF 2KV oil can. This should give me a bit of juice to fire up the 810s. I have 8 K ohms of load which I'll use for a mod transformer load, and we'll see what happens. Various AM Gurus suggest feeding the HV to both sides of the Mod transformer to keep it happy.

Probably  I should convert this to  a parallel 813 Rf Deck with a pie output, but, I have most of the parts from the previous design, so I will stick with it. For now that is..... I have some 8000s which might make a nice 20/10 rig for the upcoming solar cycle.

D,
Please be kind, as I suffer for my art.

klc
Logged

What? Me worry?
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #6 on: October 29, 2021, 12:50:01 PM »

You can get some more juice from going with a CLC power supply. Get Duncan's Amp Tools and try out your PS design there.

I have some oil can caps that you could have for a song. I have a 8mfd at 3K one I did not use on my 813 rig from the Gates iron I obtained. May be just right for the first C in the sequence. Duncan can tell you true...

I made my PS variable by using a 15A 125V variac on one leg of the primary. If yours is a 220 primary, that works really well.

Good idea to use the same HV on both Mod and RF. Safer on the mod trans. If you lose HV on the RF deck, it will also go away on the Mod, saving the Trans.

Spark gaps! See my pic on QRZ. Look for red standoffs...

Use a few strands of coax shield for a fuse for the HV. Mine is 3 strands of RG-213 shield suspended between 2 ceramic standoffs. Has not blown yet in 1yr of operation.

Don't forget a step start. A resistor and a relay works FB.

How about this - https://www.ebay.com/itm/324690117876 for your output link.

8000's in the rf deck work nice. No screen to mess with. The 8000 is the brother to the 810... HamTips has some good discussion on the differences.

If you decide to go with the 813's anyway, I strongly suggest that you wire the grid with a -75V PS plus a grid leak. Provides safety bias in the loss of drive. On the screen, use a screen dropping R. I did and is works FB. Much easier than the choke approach. I shut off the HV on RX so no wasted heat on the big R then. Math is easy. (HV-400)/80ma = R. e.g. for 1500V the R will be 13,750 ohms. Wattage is 88W. Use a 100W ceramic Ohmite/Clarostat tube R. Although, using the screen R removes some of the nice from a variable HV.

Logged
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #7 on: October 29, 2021, 01:50:55 PM »

That trans looks like it is for the Amp Supply LK-500. I don't understand how you are only getting 1kv. You should be getting 1.4 * 2135 minus whatever the choke wants to regulate it with a bridge. I would guess somewhere around 2200-2300V. That feeds 3-500's nicely which is what that amp had.  Too bad it does not have a CT. You may really want to do the variac. You then can use anywhere from 1500V on up.

Here is a pic of the LK500 PS - http://user1623492.sites.myregisteredsite.com/sitebuildercontent/sitebuilderfiles/power-supply-schem.pdf
Logged
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #8 on: October 29, 2021, 03:14:05 PM »

You could try these for your bridge. I just bought some...

https://www.amazon.com/gp/product/B087PRYGRT/ref=ppx_yo_dt_b_asin_title_o07_s00?ie=UTF8&psc=1
Logged
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #9 on: October 29, 2021, 04:07:49 PM »

J,

Yes, it's the same transformer used in the LK 500.  I want to run the modulator system at a reduced voltage for a while. Later, I'll bring her up to full V.  This thing hasn't been fired up for ~14 years; I haven't put the 810s under HV at all- today I'll hook up the mod tranny with a  6k+ ohm load and see what happens at 1 KV. ( maybe i'll try the cut wire spark gap method for insurance).

The power supply  'bridge' is on some cheep perf board, and the four 1n4007s  are only good for 4KV piv,so it will do for now.  I'm going to see if i can source locally  fiberglass board to make something more substantial with the drill press. If not, I have some thin plexi, or grab some thicker stuff from the Homely  Despot. The stand off insulator fuse is a good idea. I'll probably grab some microwave fuses later.

I have a 21.3 uF/ 5KV oil can, which I should put to use. I do appreciate the offer of the 8uF. Today, I'm using a 10uF 2.5KVcap; 930V DC across 24 K Ohms /76W load.

The RF deck will get started in a week or two. That baby will be the time killer. I have done some preliminary work  on the metering/filament/screen supply, along with layout in the cabinet. I do have a few  2 A powerstats handy, and a timing ckt for the step start. So I've got stuff to do. For now,I'll be using a B&W TVL base for the swinin' inductors. The base looks OK ish, but the other coils are suffering with polystyrene disease.

Those diode bricks look nice.

Thanks fer the input,
klc
Logged

What? Me worry?
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #10 on: October 29, 2021, 04:20:23 PM »

I used Lexan on my diode support. Should be able to get a scrap at HD. can also get small cuts on Amazon.

So, are you running 1kv by using the 230V windings on 120v? That will get you there at 1kv. Good idea.

1n4007s are a bit light. I believe that in a bridge you still have 1000V PIV, not 4K.  You may pop one on startup. 
Logged
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #11 on: October 29, 2021, 04:48:10 PM »

I've 16 of the 07's, 4 per leg. Buy 'em by the 100's get good price. I'm down to 40 of the 1n5408s; these are 3A at 1KV PIV. Don't want to waste them. I have a scrap piece of lexan or whatever it is. I would prefer pc board or G10 fiberglass, 'caus it looks so nice.

Yes, I series up the windings for now.  I'll probably put a switch on the front panel to control the 120/240 aspect; comes in handy for testing, or so I'm told.

The 8000s have a bit lower capacitance where it counts, and I got 'em. Should pick up some sockets fer the tubes.

klc
Logged

What? Me worry?
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #12 on: October 29, 2021, 05:52:53 PM »

I stopped buying anything but 6A10 diodes for power supplies years ago.  Last order was 10A10s.  They are a beast of a diode, 10A at 1kv.  Sitting in storage now, I'd of sent you some for free before my move to KP2.

With the primary series up and running on 120v a step start may be a moot point....  However a resistor and relay are cheap.

Looks like a fun project.  I already miss building but hopefully I'll be bringing the shop to me.....  Just have to find a house with a hurricane proof shop!

--Shane
KP2/KD6VXI

Logged
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #13 on: October 29, 2021, 07:16:41 PM »

Data Point - I am running my 813s at 2kV in parallel. Get 550-600W out with 750W input (2000*.375).  I can't push it harder because my plate load cap is only rated for 5Kv. Still, plenty of power.

First time on the air I caught my antenna on fire. Was a DX Engineering doublet kit fed with twinlead. Had to stop and go build a better one. Now it is just fine running a fan dipole with rg213. QRO stresses stuff a lot more that 100W.

Sounds like you are well on way to the Rally. I agree a step start is moot. I thought you were using 1 diode per side. 4 is FB, for sure. Fire it up!

How are you going to drive it?

John
Logged
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #14 on: October 29, 2021, 08:31:04 PM »


I'm plan to put a 240 V line to the basement shack sometime this year, and run things off of this  (I said this last year). Somewhere in the back-stock there are a few of the Amperite delay relays in-a-tube which  are planed for the 813 power supply. I'm finding things all over the place...

S,  Yea,  the beefier diode... The 10A10s are $.31 per 100 at mouser. Sounds like a purchase.


J,

 Drive it? it needs a ride?... I could use a DX60, or I have an Icom 706. The 706 has 160m, where the Heathkit doesn't. Or a Johnson Viking 2.

thanks guys,

klc
Logged

What? Me worry?
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #15 on: October 29, 2021, 09:28:59 PM »

You can't beat the 6 and 10A10 diodes.  They are pretty indestructible, short of a short Smiley ....

As to the step start I gave up on the Amperite relates quite awhile ago.  In private conversations with W8JI I found the Ameritron method much easier in practice.

Put 10-15 ohms in series with the primary.  Contacts wrapped around that resistor to short it out when step start 'clicks'.

The coil is put on the transformer side of the resistor to neutral.

As the capacitor bank charges the voltage drop across the resistor lowers.  This increases the voltage the coil of the step start relay sees.  At minimum Vdrop, the coil will latch and the step start closes.  Easy Peasy.

If you want protection for the resistor in case something goes wrong, wire a small 5ish A fuse in series with the step start resistor.  If you are in to lighting effects, replace the resistor with a larger incandescent light (just don't go too large or the relay won't reach V to latch).  You can watch the cap bank charge now Smiley

Hard to argue with a 2 or 3 component step start.


--Shane
KP2/KD6VXI
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #16 on: October 31, 2021, 02:34:59 AM »

Do you have a schematic of the modulator? I like the 6CA7s driving 810s and wonder about whether it uses any negative feedback to keep the signal linear over the half cycle when the 810s draw grid current.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #17 on: October 31, 2021, 04:54:32 PM »


A quick post as I'm at work....

I wired everything up last night and got the Mod transformer to talk back to me.The scope shows a nasty 120Hz hummm across the modulator load, but at least I'm getting something to work with. Maybe more later tonight.

KLC
ps    i've a partial  " as found  "   schematic that I'll look at tonight.
Logged

What? Me worry?
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #18 on: November 07, 2021, 12:54:53 PM »



Update.

I spent the last two days playing with DigiKeys Scheme-it  to generate a usefull schematic to post here..... It didn't upload to digikey, so its sitting on the home computor. I intended to print the schelmatic  here at work, scan and upload to da fone... no joy, so I'll try another way....

I don't like the 60 V p-p sitting on my power supply load, so thats gunna be the fun thing to do. BTW, the mod deck works with the low ~ 1KV on the 810s, but the ripple makes me sad.

33 uF total after the choke; dont know the Heneries, yet.

Mo to follow.

KLC
Logged

What? Me worry?
W7TFO
WTF-OVER in 7 land Dennis
Contributing
Member
*
Offline Offline

Posts: 2521


IN A TRIODE NO ONE CAN HEAR YOUR SCREEN


WWW
« Reply #19 on: November 07, 2021, 01:08:13 PM »

Kev, you may have an iron deficiency.

Your project may have caught it from you.

If a heavy cure is needed, ring me up. Or some such, many cures for such lackings here.

73DocG
Logged

Just pacing the Farady cage...
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #20 on: November 07, 2021, 03:23:22 PM »

KLC,

  The ripple on the B+ is an RMS value, not peak to peak. As an approximation, 60v P-P is 30v peak, and .7X that for RMS, = 21v. Yes the ripple is not sinusoidal, so this is an estimate only.
Then (21v/1000v) = .021 Then take that X 100 to get percentage, or 2.1%
The B+ supply for a plate modulator, or RF final at 5% ripple or less is usually fine. You are way better than that. If you want better than that, add another choke, and another 33 uf for a L-C-L-C filter. Tongue

   A Push Pull Plate Modulator should null out the power line ripple from the modulator B+ so long as the tubes are balanced and driven equally. The 2% ripple on the RF PA B+ will modulate the RF 2% at 120hz, but most BA receivers do not reproduce frequencies that low unless they are "souped up". With a Flex, or other SDR, I routinely see the old BA transmitters with +/- 120hz sideband energy, but the level is usually quite low, and usually not audible. To me this is a signature of a vintage rig, and not necessarily indicating a problem.

Jim
Wd5JKO
Logged
VE7RF
Member

Offline Offline

Posts: 56


« Reply #21 on: November 07, 2021, 08:46:22 PM »

I stopped buying anything but 6A10 diodes for power supplies years ago.  Last order was 10A10s.  They are a beast of a diode, 10A at 1kv.  Sitting in storage now, I'd of sent you some for free before my move to KP2.

With the primary series up and running on 120v a step start may be a moot point....  However a resistor and relay are cheap.

Looks like a fun project.  I already miss building but hopefully I'll be bringing the shop to me.....  Just have to find a house with a hurricane proof shop!

--Shane
KP2/KD6VXI



Shane, are the 10A10  diodes the same diam and length as a 6A10 ??   Does the 10A10  use the same  16 ga wire leads on each end ( .051") ?
 The surge is the same on both... 400 amps.
There is another diode, called a 10AO8, which is rated for  1kv  @  10 amps... but has a 600 amp surge rating.
On all those diodes, they require 1.1" square inch of surface area on a  pcb  if run at rated average current.....and  2.2" square inches  at every series junction.

A while back, I was experimenting, using 49 x 6A10's on perf board.... for bias for a 3x3 tube.  It was in a test jig, with a 0-60 vdc lab supply..and a series
  25 ohm / 100 watt ww  resistor.   The  6A10's  ran pretty warm ( not quite hot)  with  2 amps  CCS flowing.    The  1N5408's ran pretty warm ( not quite hot, but getting there)...with  1 amp  CCS flowing.   I noticed the V drop increased by  12%..when increasing the current from  200 ma..to  1/2 amps.

I noticed an anomaly though.   The  1st and last diode ran stone cold.  ( diodes  #1  and  #49).   Diodes 2-48  were getting borderline hot.   With 6 amps CCS flowing through a 6A10, it would burn my fingers.   6 amps X .78 V  =  4.68 watts..so no wonder.

Then it clicked, with  test leads  connected to 1st and last diodes...the medium sized test lead clips are heatsinking the input of  diode #1..and the output of  diode  #49.

It's all semantic for the most part though.  The general idea is to keep the leads long..as they are doing the heatsinking.

In normal operation, like a FWD, or  FWB, the diodes are only running at a 50%  duty cycle..with a CCS load on the supply  ( cap input or choke input or  C-L-C  filter etc.   Only 2 of the 4 x legs of a FWB are conducting at any one time.  That explains why  Dahl  rated his FWB rectifier assys' at  12 amps CCS, when using 6A10's in all 4 x legs.   (he rated the  3 phase version, with  6 x legs, at  18 amps CCS..also using 6A10's in all 6 x legs).

BTW, the last batch of bulk diodes I bought were PM-600's.  These are identical to 6A10's.   I since found some cheap sources for  10A10's in bulk,  ( typ boxes of  100 or  250).

The  6A10 /  PM-600 is a good diode.   It's the same length as a 1N5408...and same  16 ga wire leads. The 6A10 is double the diameter of a 1N5408.  Point here is... a 6A10 /  PM-600  will drop right in...where a 1N5408  once stood. 

Later.... Jim   VE7RF
Logged
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #22 on: November 07, 2021, 09:29:03 PM »


It's late, but, in the interest of Science.....Some pics and some actual numbers.

The load is 6 k, I used a .001 / 3000V cap and 10X probe fer the scope. The Fluke is a Fluke. 830 V dc.


* Mod Deck voltage.jpg (4920.41 KB, 4032x3024 - viewed 252 times.)

* Mod Deck load.jpg (3951.37 KB, 4032x3024 - viewed 253 times.)
Logged

What? Me worry?
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #23 on: November 08, 2021, 05:40:41 AM »

Jim,

I use very long lead lengths on the diode banks I build. ARRL manual taught that decades ago.     When using silicon rectifiers the leads are heat sinks.

Truck is, I solder the leads together for the entire lead.  This gives more structural rigidity as well as a lot more surface area creating more heat sink.  The diodes are placed almost the entire lead length off the perf board. I do the dame thing when I used pc board material.

The 10A10 where probably the dame diode as the 10A08 you spoke of.

You have to be careful when bargain shopping for the big diodes.  I got some P600 diodes on a reel once that where 50 volt rectifiers!  Makes for a great turn in event when using to rectify 4.5kv ac!  Not sure why, but they looked identical to 6A10s, same on the diode meter, but after looking up the different numbers on the side of the reel, they where 50 volt.  Marked as P600, which cross referenced to 6A10.

So I then had a ton of rectifiers for bias boards.  Nothing really lost, other than that first bridge!

--Shane
KP2 / KD6VXI
Logged
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #24 on: November 08, 2021, 06:08:53 AM »

For 10 bucks you can get these on amazon...

https://www.amazon.com/gp/product/B087PRYGRT
Logged
Pages: [1] 2   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.068 seconds with 19 queries.