It's jumpered for 120V.

I am reading 120V on the output of the soft start.  It's installed according to the Harbach instructions.  But I haven't puzzled out how to tell if it is coming through the resistors or the relays.  

As much as I don't want to do it, I think I need to de-solder the fan leads and connect them to a lamp cord with some wire nuts and just see if the fan works at all.  If it does, that's one easy box to check off.  

When my HV probe gets here, I can test the plate transformer's secondary.  I'll have to wait until the rest of the family is gone and lock the cats out of the room so they don't jump on the amp.  

Looking at the schematic, it looks like when the amp is wired for 120V, AC can pass from the terminal strip through the primary windings of the filament transformer and appear on the side of the power switch that is open when in the off position. So maybe I shouldn't be surprised to see 120V there.  

You can follow the path from terminal 3 (which connects to terminal 4 for 120V operation) down to the lower winding of the filament transformer primary, then up to the open side of the power switch.

Terminal 2 of the strip connects to terminal 1 for 120V, and it goes down to the upper primary windings and back up to the open side of the power switch on the second set of poles.  I get 120V across those open side power switch poles when the amp is plugged in.  

The soft start module is spliced in to the lines coming from terminals 1 and 4 and the output goes to the hot side of the power switch.  Power for the relay coils on the soft start comes from the rectified output of the lower secondary windings of the filament transformer (second lead from the bottom of the schematic to the right of the transformer).



Attached File

Yeah I am going to check out the fan and see if it works connected directly to a lamp cord, and then step back from it for a day or two until my HV probe shows up and I can get to town for supplies to build a Jesus stick.  If I want to include a resistor in the stick, that might require waiting on shipping since there aren't any local places that sell them after Radio Shack closed.

I have had the same experience with other puzzles.  Sometimes it's easier to solve them when you aren't actually staring at them.

1. Both. Metal surfaces oxidize HORRIBLY with RF and age and its worse at higher frequencies. Its due to Skin Effect. Disassemble as much as possible /if possible, down to the brass shaft springs (should be leaf springs)
Use Brasso and a Q Tip. If they are soldered in, burned, worn etc, then they cant be fixed.

Its better to "ground' (I hate that word, "chassis/common" is more accurate) the rotor SHAFT. Thats what I do in the Drakes. Use some kind of brass spring to press against the shaft and ground the spring.
It works better than original

2. This indicates an open-circuit down the line. This is a problem with our modern, very high resistance Multi Meters, usually 10 meg ohms or more. If theres leakage resistance thru the switch, imagine its now a resistor. If that leakage is much less than the 10 Meg Ohm resistance of the meter and the down-stream circuit is open, the meter will "read through" the switch. Volt meters do not measure voltage, they measure small currents.

If you really get hung up bad, Email me. Ive done an excrement - load of such work over 40 years...

3. Great its been thru a Butcher Shop...Fortunately RF power amps are rather simple circuits, its just very confusing if you dont have much experience with it.

Get three Harbor Freight multi meters, they are (or were) rated for 1 KV each. String 3 in series. Works great, just use HV insulated wire.

$30.

also... get rid of those strings of rectifier diodes, thats a DISASTER just waiting to happen if theres a transient. That power supply rectifier setup is WRONG. A diode string must have equalzing caps across the diodes to shunt pulses that can cause one or two diodes to turn on before others. That results in excessive voltage across others and they break down.

If the PS breaks down, it can sink high currents and destroy a transformer. $$$$$$$$$$$

1N4007 are good for 1.5 KV PRV. Better yet, get two commercial rectifiers (several KV and an amp or two). I just built a 2 KV supply from a MOT and used two, 1N4007s in series for each leg of a bridge.

I built a big DC supply and used 8 KV, 5 A commercial rectifiers, absolutely bullet proof.

They were ONLY $15 each. But for what youre saving in not buying a new one, you can justify it.

This is incredibly poor advice.

Then the SB220 rectifier circuit has been "wrong" for 40+ years?

Also, how do you propose to fit you "better" rectifier in the current case design?  

[Deleted]

[Deleted]

[Deleted]

[Deleted]

[Deleted]

[Deleted]

[Deleted]

Hey guys, this is a tech forum. Not the "mine is bigger than yours forum".
Please do explain your reasoning why the rectifier circuit has been wrong for however long its been built, but save your put downs for somewhere else and try and help the guy out (like has been mostly going on).
Thanks.

Seems weird to me that someone would join a gun forum, find a sub forum of a sub forum right away. In fact, the only posts are in that particular forum, and contain words that seem to be designed to rile people up.

Wow, I go have a few drinks with a friend and I missed a bunch of drama it seems.

I'm sad that my thread took a bad turn.

That is ok, looks like it is back on track at this point and many here are still following it. So please continue hoping you get this thing running at some point.

my ignore list again turns out to be 100% accurate.

op

do you have the original healthkit build manual?

I have the one that was used to build my SB-220. if you need a build manual I can scan mine and email it to you.

I was going to sell the manual with the amp yesterday but got lazy and never set up a table at the hamfest. So I still have it.


ETA: You probably already have one, I googled and it is readily available online in pdf format.

I downloaded the sb220 assembly here:

https://www.rsp-italy.it/Electronics/Kits/_contents/Heathkit/Kits/Heathkit%20SB-220%20Linear%20amplifier.pdf

I'm glad the troll is gone, let's get back after it!

I was going to ask him last night if this was the last ham forum he had not been banned from.
Let's get back to the really cool resto project!

I eagerly await his next Youtube video about how terrible and wrong we all are.

Yes, thanks for the offer.  I do have a .pdf of the manual.  It's more than I want to print in it's entirety but when I get it working I will print the operating instructions to have on hand.  I've been referring to the .pdf for now.

There are a couple of redrawn schematics floating around out there that were done by Tom Hammond N0SS (sk).  One of them attempts to correct an apparent error where Heathkit makes some connections to the transformer before the breakers.

Thanks, I'm glad it didn't have to result in a thread closure.

Okay so just to simplify things, I removed the soft start.  I also removed the wires running to the power switch, which were green in my amp, but are black finally of the reference materials.  These I am replacing with 14 gauge 600V stranded hookup wire.

While I have it apart, I re-checked the power switch for open and close operation on both sets of terminals with my continuity tester. It functions properly.

I also cut a lamp cord and connected the fan with wire nuts.  The fan works!  This helps narrow down the problem.  

Also my HV probe showed up today.  I'm not sure I want to mess with the RF deck yet.  For now I've numbered the terminals on the voltage selector terminal strip and I'm going to backtrack through the Heathkit manual and make sure all of the connections from the breaker forward to the power and CW/SSB switches are correct.  With AC, you would think it shouldn't matter, but maybe it's an "out of phase" condition or something, assuming the plate transformer isn't bad.  Is it possible to connect different primary windings out of phase and cause problems?

Attached File

Success!!!

After removing the soft start and rewiring the 120/240V selector strip, I show HV on the meter, the meter bulbs light up, the fan runs, and if you look at the top right of the photo, that orange glow in the fan area is from the tube filaments!

Attached File


Attached File



This means both transformers are probably okay.  I didn't try the CW/SSB switch yet.  I am too happy to see things running to want to mess with it right now.

So this means that either the soft start circuit was causing a problem, or it was a wiring problem with the 120/240 voltage selector strip.  

Now that I know it works at least partially, I think it will be worth it to invest in new parasitic suppressors and to take the time to ground the grids.

 I'll need to dig through some junk boxes and find an RCA cable to test the keying circuit and relay.


Very happy today!

Very nice super happy for you. I really like the SB-220 amps

Thanks!

I really appreciate yours (and others) suggestions and advice in this thread as I've worked on this.

Looking ahead to the future, I think I might eventually buy some modern, commonly available switches and design some 3D printed bezels that will fit the holes in the front panel.  

In all the reading I've done about these amps, it looks like the meters and the switches both are almost impossible to obtain.  I can't really do anything about the meters, but it would help to have a way to mount some newer rocker switches that filled the holes and made them look better.

I'm fortunate that my switches seem to work okay.

Wooooohoooo!!!  Let the amp set on for several hours before hooking the high voltage to the tubes, this it to regetter them, then run on CW/tune in normal operation for an hour or so before switching to SSB.  If a tube flashes over, it's junk.

I still need to buy or build some new parasitic suppressors and a keying cable before I can run it on the air.

It is all back together with the cabinet on, but with the plates disconnected.  

I probably need to oil the fan bushings as well before letting it run for hours, but it might be hard to get to without some minor disassembly again.  I should have done that before putting it back together but I didn't think about it.

I did test the CW/SSB switch and it does go from about 2.4 to 3kW when I flip the switch.

I also checked the 90V secondary for the keying relay on the filament transformer and it reads 88 volts on my meter so that looks good as well.

So, it still remains to be seen if the tubes will work well and if the tuning capacitors will work.  Also the band switch was corroded black.  I did use some contact cleaner on it.  If all that works, I am only out the cost of some Harbach kits, some wire, solder, new rubber feet, some crimp terminals, a few nuts and bolts, and that's it.  If you count the test equipment and tools I bought, then you can add more.  But those will come in handy for other projects.

If everything does work well, I might go to the trouble of re-painting the cabinet and getting an overlay for the front panel.  I might try to find an unblemished knob to replace the one with the ding.

There isn't much that can be done for the scratched meter faces.  I already cleaned them up as best as I could with some plastic polish.

In the future I might upgrade to LED meter lamps with a green hue to match the front panel.  And I might give the soft start kit another go once I am sure everything is going to work.

I could also paint the transformers to hide the rust, but it isn't really hurting anything and it's hidden with the cabinet on.  I've also considered buying a new power cable since the one that's on there now is grungy and my attempts to clean it have failed.

There are other mods I could do.  There is the "series glitch resistor" mod, and the Amp Repair Guy who is taking over Harbach recommends replacing the plate blocking capacitor because the original one isn't rated for RF.  I did do the bias mod on the relay while I was in there.

Anyhow thanks for your interest and suggestions.

nice job

I dont remember if you mentioned changing the keying circuit or not or if you are planning to use a keying interface, but you have to do one of those because the SB-220 uses 120 VAC in the keying circuit and will fry the keying circuit in a solid state transceiver


also be careful with the band switch, they are unobtainium. When i did the harbach power supply caps and diode board, the instructions called for taking the front off the cabinet but that puts that band switch with only one end attached, so I found a way to replace both parts with out taking off the front of the cabinet and keeping both ends of the band switch attached

here is a pic of mine

Attached File

Yes, I installed the Harbach soft key kit so I can use it with modern radios.  

Harbach did have the input bandswitch wafers but the price on them is crazy.  They don't have the output wafers.

Without them, some other way of making the proper connections would have to be devised.  Someone smarter than me could figure out a way to use a regular rotary switch and a bank of relays to make the proper connections.  It would be expensive, but with the price of the wafers, it might be a break even solution.  There's plenty of room in that compartment for some relays on the input side, and you could probably fit another board on the other side, and pass a ribbon cable through the hole currently occupied by the bandswitch shaft.

I see you have the standby mod.  I read about a guy who also set their fan speed up to reduce by half when in standby to make it quieter.  Nice looking amp!

I like the standby mod. I need a light in my eye to remind me it is still on LOL. Plus I like to bypass the amp when doing initial tuning. That mod makes it easy.

but I bought it that way, I didn't do the mod. The only mods I did was the supply cap board and the diode board with meter protection both Harbach mods. It had everything else when I bought it. I think it has a keying mod, but I didn't want to trust it since it looks homebrew. It also has a soft start mod, that also looks home brew.

I'll have to look into the standby thing.  I really like to keep the look original though, so if I did the mod, I'd probably try to find a way to activate it without drilling the front panel for a switch.  Maybe a magnet switch or something hidden like that.

There's all kinds of things you could do using multicolor LEDs for the meter lights.  Make red mean TX, green RX, and maybe orange or blue for standby.  If I am ever in a position to retire while I am not yet senile, that might be a fun project.

It looks like all you need to do for standby (pass through) is to disable the keying circuit.  You can even get a box with a switch in it and RCA in and out.  This would be helpful for adjusting antenna tuners without putting an amplified signal out there.  

I just ordered a Palstar tuner from DX Engineering that has an integral watt meter, and my dummy load showed up.  I probably need to check the keying circuit so I can order a new T/R relay if need be.

Also I ordered some nichrome wire for making parasitic suppressors and some copper strip material for grounding the grids.  That will involve partially disassembling the amp again to be able to drill the holes and get the screws in.

I still need to get resistors for the parasitic suppressors.  Not sure what to order as the original carbon composition ones used a single resistor, but the new U-shaped suppressors I see online use 2 resistors in parallel.  They must be a different chemistry but still non-inductive.  I also learned it may be difficult to solder to the nichrome wire without acid core solder.

The guy in this video is occasionally a bit controversial, but the information contained in this video is good and will get you to a good set of parasitics.

https://youtu.be/4e9lj8OTPFE?si=cNepnJDxgwYtq0Pt


Copper strap isn't necessary to ground the grids, I used some 14gauge solid for mine and used existing holes in the chassis to ground them to.

I've read about it both ways (strap and wire) and seen it both ways in videos.  From what I gather, the tube's pin leads and the sockets themselves have a self-resonance that contributes to the parasitic oscillation.  Flat straps and short, direct paths to ground for each grid pin are supposed to help reduce this problem.  That's supposedly the reason Heathkit has chokes between the grids and ground, to act as filters for this self resonance.  At least, that's what I gathered from my reading.  The problem is that one of the chokes or caps failing can cause damage elsewhere in the amp so that's why they recommend the direct grounding.

I do like the simplicity of using round wire or flat braid for the grid connections, and it would be nice to skip drilling additional holes.  As I understand it, the more effective the parasitic suppressors are, the less of a concern the round wire is on the grid sockets.  If I were using the factory suppressors (even if new) the type of grid connections would matter more.  I'll probably land somewhere in the middle with copper strap for the connections, but not drilling additional holes.  

One of the issues I have is the builder of this amp soldered the 2 capacitors on the little terminal strip that holds the filament power wires and choke on the wrong side of the strip, making it difficult to get to the nut, which is also holding the solder tab for one of the grid capacitors.  They did hook all of the leads through terminals everywhere, so desoldering and changing anything has been laborious.

The nichrome wire is supposed to have 10X the resistance of copper, so they recommend using it for the suppressors. It's supposed to help reduce the Q and help the suppressors be more effective.  

Any updates on this?  Is it amplifying RF?

Not yet.  I've been waiting for some stuff to arrive that I ordered.  I'm putting in a HV fuse and series glitch resistor.

I also want to ground the grids.  In doing so, I needed to remove the big coil on the underside of the chassis.  The builder soldered the two ceramic caps in backwards blocking the nut, so I had to do a lot of desoldering.  After putting it back together, I'm not happy with how it turned out so I am going to desolder and resolder it again.

I did get some Ohmite OY series resistors which are a modern non-inductive replacement for the original carbon composition resistors.  I'll use these to make new parasitic suppressors.  I chose the same values, 47 Ohm, 2 W.

And I also did a lot of reading and watching videos on gassy tubes.  Apparently the 3-500Z can't be de-gassed just by heating the filaments. It's something about the gettering material being applied to a different part of the tube.  They need to be run hot enough for them to glow a dull red.  The suggestion is to set them up as a diode and run the filament on a transformer and then use a variac through a rectifier to apply enough voltage to get around 300-400 milliamps flowing through the tube.  

Regular operation is supposed to keep them from getting gassy.  It's all anecdotal, but apparently 50 year old tubes that are used occasionally tend to work fine, but new old stock tubes that have sat unused for decades tend to be gassy.

Anyhow I ran out of time to finish the soldering work I started.  I hope to get the grids done today after work.  All the capacitors and the two chokes are out now at least.  Desoldering them all was a lot more work than I thought it would be, mostly because of the way they hooked the leads through everything.

Grids grounded finally.  I drilled holes in the tops of the straps where they contact the socket tabs to make solder windows so I could be sure of continuity.

Attached File



It looks like the lead going to the plate blocking capacitor is made of hollow tinned braid, and they just slipped it over the suppressor leads and soldered it.  Almost zero chance I'll be able to get it off there and be able to reattach it to new suppressor leads the same way.

Attached File

You should be able to take a HOT 100 watt iron and a solder sucker to that and get it apart without melting anything important.


Edit:


The braid is on the second tab on mine.

Well I managed to get it back together.  My wife was afraid to be a third hand so I taped the suppressors so they wouldn't slip while I soldered the braid on.  It's ugly but it will work for now.  I have new 4mm braid on the way and I'll replace it when I get the fuses and the series glitch resistor here.

 After removing the coils, I tested the old 47 ohm resistors and they tested 53 Ohms, so not too bad and they probably could have been left in.  But these new resistors are the correct value and should hold their value much longer.

Attached File

So today I replaced the old braid running from the suppressors to the doorknob cap with some new 4mm tinned braid so it doesn't look so sloppy.

While I was in there,  I disconnected the anodes again and I'm going to give it a few hours with just the filaments running.  I've read that this doesn't do much, but anything can help.  Good news is we are at somewhat higher elevation 6600' here so less air and lower air pressure means there probably would be less contamination than the tubes might see at sea level.

I've also read about a method for running the amp de-tuned into a dummy load for de-gassing, and I might give that a shot later if I can find enough info on the process to feel confident I wont screw something up.  

I am still waiting on a fuse block and resistor to ship, but I might take a ride into town later and see if I can find something at the auto parts store to save me a little time.

Attached File

I've been reading that (careful) intentional mis-tuning can be used to de-gas these tubes if you don't let it get out of control.

Eimac says running the filament for extended periods will degass the tube, lots of others say the anode has to be cherry red to degass.  The detuning is a little risky, just keep an eye on your grid current, or you could tune it up on AM in the CW voltage position and call CQ with around 60-80 watts PEP drive for about 5 minutes and do the same thing.  If the tube doesn't flash over when high voltage is initially applied, you should be good to go.