• You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.
  • Click here to find out how to win free radios from Retevis!

Maco 300 power lag

Cable Guy

Packaged by weight, some settling may occur.
I Support WorldwideDX.com!
Dec 29, 2010
447
370
73
West Tennessee
I have 2x Maco 300 amps. One is a 7 tube, the other has 6 tubes, with the 5th final tube removed.

They both exhibit a strange issue where dead key, after keying for a short time, starts to drop from about 250W down to 100W. After a little while, it climbs back up to regular output and stays for a while. This lag in power also affects peak output. I have tried other radios, antennas and dummy loads.

Thinking it could be the 'lytic caps, I have replaced all. It still occurs.

Has anyone experienced this and if so, what caused it?
 

I have 2x Maco 300 amps. One is a 7 tube, the other has 6 tubes, with the 5th final tube removed.

They both exhibit a strange issue where dead key, after keying for a short time, starts to drop from about 250W down to 100W. After a little while, it climbs back up to regular output and stays for a while. This lag in power also affects peak output. I have tried other radios, antennas and dummy loads.

Thinking it could be the 'lytic caps, I have replaced all. It still occurs.

Has anyone experienced this and if so, what caused it?
Neither the screen Grid or control grid voltages are regulated on this amplifier. The screen grid looks to have the same positive voltage that feeds the relays and is not likely to be moving around. However that method of a resistor and diode or just the resistor on the final stage to establish a control grid voltage is most likely the one that's moving around on you. As the tubes age they tend to develop more secondary emissions and that moves these voltages around as the tube gets hotter.

The control grid is self biased through a resistor in the final stage and dependent on how much current is being drawn to establish the voltage. The driver seems to have a diode rectifying RF to produce its voltage. A much better method would be to have a power supply at a zener diode to regulate it's bias voltage.
 
All good advice, but I'd be looking for a heat-related fault. As in, a capacitor that gets hot and stops performing.

The blocking cap between the plate choke where the tube-cap clips all connect on one side, and the Plate Tune control on the other side can do this.

Too much voltage there to just lay your fingertip on it to see if it's hot enough to burn you.

If that part is causing the problem, you'll see that the peak position of the Plate Tune knob has shifted when the power falls off. If the Tune control's peak position does NOT change when the power falls, this part is not likely the culprit. An infrared video camera would reveal this, but only if you have one.

73
 
  • Like
Reactions: Cable Guy
I will check the tune control when it happens, and see if it peaks at a different position. I have a thermal scope that doesn't give the temp or change color, but does change intensity as the part gets hotter. I use it sometimes to find heat related faults in radios, but never thought to use it here. Thanks for the responses guys.
 
The schematic states .006 for that cap, I guess that's in uF? Is that an NPO ceramic? It's not green like other NPO I've seen. It's rather large as well. I figure voltages to be 900-1200 vDC. I need to buy some for these amps, and some for a stash. I see some caps listed that look smaller, and I know the electrolytic have gotten smaller since the 70's, is it the same for these ceramics? Could you point me to a seller?

I'm sorry I have so many questions, but these tube amps have always been a gray area for me and I know this is the place to sit and color the blank pages of my gray matter. If only I could stay inside the lines...
 
A capacitor marked ".006" is microfarads. Yeah, it's also six nanofarads, but it seems only geeks use that unit in the USA.

It's also 6000 picofarads.

Here's what I would recommend. The RF current is pretty high in that spot, and a capacitor's RF resistance is a big deal. Simple rule. More metal equals less resistance. Voltage rating needs to be double or more the tubes' plate voltage. Triple is safer. Finding a capacitor's current rating isn't as easy as it sounds. Bigger is just better for this circuit function.

This one looks good for a sweep-tube amp.

https://www.rfparts.com/capacitors/caps-highvoltage-plate/004-3kvhda.html

That's what I would go with.

73
 
All good advice, but I'd be looking for a heat-related fault. As in, a capacitor that gets hot and stops performing.
Look at the symptom again...

"They both exhibit a strange issue where dead key, after keying for a short time, starts to drop from about 250W down to 100W. After a little while, it climbs back up to regular output and stays for a while."

A capacitor failing from too much RF current will not drift out of tolerance from a little heat, and then drift back into tolerance after being keyed for a longer period of time. If it continues to drift with more heat, it continues to drift in the same direction.

These are more likely the symptoms of weak tubes, developing secondary emissions. The weak tubes fall flat when the cathode can't sustain the current. As the tubes get hotter, they will develop secondary emissions that drive the current back up. Measure control grid and Screen grid voltages. They are most likely moving as the power output is shifting.
 
Measure control grid and Screen grid voltages.
Or just swap tubes. I have looked down into the top mica insulator on the tubes in more than one linear only to see the screen grid begin to glow red, then darken when it goes back to receive mode. Typically the result of the Load control being too loose, maybe? Tubes that got this kind of stress may well be the root of the symptom.

73
 
Look at the symptom again...

"They both exhibit a strange issue where dead key, after keying for a short time, starts to drop from about 250W down to 100W. After a little while, it climbs back up to regular output and stays for a while."
I clung to the capacitor fault idea because while power is reduced during its fits of failure, the tune control does see a peak in another position and when the fit is over, must be peaked back to original position.

I will also check the grids per your suggestion.
This one looks good for a sweep-tube amp.

https://www.rfparts.com/capacitors/caps-highvoltage-plate/004-3kvhda.html

That's what I would go with.

73
Thanks, I did order some to try. I often see them in there, and knowing the pressure they're under, really not sure why I haven't seen more of them failed outright. I have only seen a handful of failed ceramics, and most of those were failed from physical damage. Maybe I have overlooked them too often.
These are more likely the symptoms of weak tubes, developing secondary emissions. The weak tubes fall flat when the cathode can't sustain the current. As the tubes get hotter, they will develop secondary emissions that drive the current back up. Measure control grid and Screen grid voltages. They are most likely moving as the power output is shifting.
This, I haven't ignored. I'm seeing it in my mind. So what I gather, the grid/bias circuits could be wonky, the tubes could be weak and going into SE, or the final bypass cap is going thru changes, or possibly some combinations of those. Now I have plenty to do. Thank you all.
 
  • Like
Reactions: Naysayer
Id like to see a meter on the input side looking at the input tune. If the filament choke is not sufficient or up to snuff you will get a symptom as you described. The input tune will look like its varying and be unsteady. Or could be bad 50 year old tubes.
 
I remember years ago reading a book about television servicing. It was written in the 1950s. The advice was about working without a schematic diagram. He suggested that probabilities are a good guide if you lack a fault chart or proper diagram. Said the most-likely part to fail in a 1950s TV was a tube. Second-most likely was an electrolytic capacitor.

Some things never change.

73
 
  • Like
Reactions: Cable Guy
Another thing I just thought of is the shifting output power alone, regardless of the reason for it, is going to cause the resonant load impedance to change, and therefore the amplifier will tune in a different spot anyway. The only way to prevent a shift in output power from changing the resonant load impedance, is to do it by adjusting plate voltage. That voltage is not likely to be changing here.

We also have to remember that the DC blocking capacitor is not part of the tuned circuit, as though it were in parallel with the load or tune controls. Its value is not critical at all and can drift considerably before impacting output power. Normally when a plate voltage blocking capacitor fails to the point where it impacts performance, there are physical signs of it.
 
Last edited:
  • Like
Reactions: Cable Guy
Another thing I just thought of is the shifting output power alone, regardless of the reason for it, is going to cause the resonant load impedance to change, and therefore the amplifier will tune in a different spot anyway. The only way to prevent a shift in output power from changing the resonant load impedance, is to do it by adjusting plate voltage. That voltage is not likely to be changing here.

We also have to remember that the DC blocking capacitor is not part of the tuned circuit, as though it were in parallel with the load or tune controls. Its value is not critical at all and can drift considerably before impacting output power. Normally when a plate voltage blocking capacitor fails to the point where it impacts performance, there are physical signs of it.
Bad Cap.jpg
Could it look like this?
 

dxChat
Help Users
  • No one is chatting at the moment.
  • @ Deon:
    Good day, I have an Icom IC910H radio. I want to program a memory with split frequency tone on TX only and NO tone on RX,. Is this possible to do from the face of the radio? I don't have the interface to connect to a computer.
  • @ crappykraco:
    @Deon very unlikely to get a response here. You are better off starting a thread on the topic.
  • dxBot:
    Backwoods40 has left the room.
  • dxBot:
    GlocknSpiel has left the room.