• 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.

1X4 with disabled driver

2NC995

DAN
Feb 23, 2008
2,219
4,005
273
Coastal NC
Ok, so I have an old “powerplate” 1x4 that I am tinkering with. The board is an absolute hack job so I will probably end up rebuilding it using the existing heat sink, case and transistors.

So far I have:

1. Disabled the driver and adjusted the input tuning.
2. Added bias per the above “quick and dirty” thread. This still needs to be dialed in, and I may decide to go with a regulated system in the long run, once I wrap my brain around it.

One thing I didn’t consider but am now realizing, is that this amp is seems to need a much higher drive level than I expected. I realize now how between the input padding and the very different setup why they’re able to get away with 1-2879 driving 4-2879s.

Anyone have any speculation as to what would be a good SSB drive level for an amp of this design (when properly biased)?

I don’t like to run my stuff hard...I cruise my DX500 at 12w in, but this thing is virtually asleep at that level.

Anyway, was hoping to run it with my Uniden 980, but now I’m thinking it’s going to take a dual final radio.
 

Chances are the input to the 4 pills and output of the 1 pill driver were not set up at 50 ohms. They didn't have to take the time to do that as long as both matched. What is the input VSWR when driving the amp? Using a simple pi circuit with a pair of 464 compression caps and 5 turns of wire around a crazy glue tube should get you close. Pictures of the existing circuit might help with improved suggestions.
 
  • Like
Reactions: Shadetree Mechanic
Chances are the input to the 4 pills and output of the 1 pill driver were not set up at 50 ohms. They didn't have to take the time to do that as long as both matched. What is the input VSWR when driving the amp? Using a simple pi circuit with a pair of 464 compression caps and 5 turns of wire around a crazy glue tube should get you close. Pictures of the existing circuit might help with improved suggestions.

Input SWR is barely moving the needle. It has a single 465 trimmer off the input combiner. The lowest SWR is with the trimmer backed off all the way. I plan to put a lower value trimmer in place to get more towards the middle of the range.

I think I read somewhere that the input transformer windings determine the drive level? Maybe if you want to use a lower drive level, you could put more turns on the xformers? I'm not sure exactly how that works.

I think you’re right. I ran it up to about 350 watts, but it took 40 watts in to get there. My DX500 will do that with much, much less.
 
Chances are the input to the 4 pills and output of the 1 pill driver were not set up at 50 ohms. They didn't have to take the time to do that as long as both matched

This appears to be the issue. I put it back into class C, and find that If I adjust the input capacitor to show highest output, then the VSWR climbs to 6 at the radio. If I adjust for lowest VSWR at the radio, power plummets.

This is the same amplifier plan that’s been spread all over creation...with the driver removed it looks just like this (but without the padding resistors).


4B2BA0FE-8D22-4EAD-A979-835F1739AC22.jpeg
 
Last edited:
It sounds like the input impedance is too low. You have a choice here to just add one turn to both input transformers to raise the impedance a fixed amount and hope to pad it into perfection with the 465. Or, you could add a coil and another trimmer cap to form a PI input matching circuit. This would allow precise adjustment with the caps and bending the coil for proper inductance and best match. This also provides harmonic reduction and improved IMD performance.
 
  • Like
Reactions: 2NC995
It sounds like the input impedance is too low. You have a choice here to just add one turn to both input transformers to raise the impedance a fixed amount and hope to pad it into perfection with the 465. Or, you could add a coil and another trimmer cap to form a PI input matching circuit. This would allow precise adjustment with the caps and bending the coil for proper inductance and best match. This also provides harmonic reduction and improved IMD performance.

Thank you. I really appreciate your thoughts on this. Struggling with the bias as well...the voltage seems to plummet under TX. That’s what prompted me to go back to C and work on the input tune first.
 
Thank you. I really appreciate your thoughts on this. Struggling with the bias as well...the voltage seems to plummet under TX. That’s what prompted me to go back to C and work on the input tune first.
If you're using the bias circuit described earlier with the 100 ohm resistor and diode, it's not possible for this bias voltage to plummet under TX. What you are more than likely seeing is the result of RF getting into your volt meter.

You may have to bypass your meter leads at the meter end with a .01 cap right across the banana jacks. Modern ones are a pain because they have the insulating sleeve around the banana pin that probably needs to be cut off first. Some DVM's even require ferrite beads on both leads.
 
If you're using the bias circuit described earlier with the 100 ohm resistor and diode, it's not possible for this bias voltage to plummet under TX. What you are more than likely seeing is the result of RF getting into your volt meter.

You may have to bypass your meter leads at the meter end with a .01 cap right across the banana jacks. Modern ones are a pain because they have the insulating sleeve around the banana pin that probably needs to be cut off first. Some DVM's even require ferrite beads on both leads.

Ok. That’s good news. It made zero sense as it was dropping to .04V, and I could not wrap my head around that.
 
  • Like
Reactions: Shockwave
Without bypassing the leads, many meters have trouble measuring voltage inside amplifiers with the presence of a strong RF field.

Ok...the follow up to that is, am I good at ~.7v for bias voltage for a pair of 2879s, or do I really need to be measuring current?

I used a diode and 50ohm sandbar per pair, with bypass caps from transformer to ground and across the diode.
 
  • Like
Reactions: Shadetree Mechanic
Without regulation and ability to adjust bias current, you should only use this bias circuit with 100 ohm resistors. Cutting that resistance in half will force twice the current thought the diodes and your output transistors. If you have four of the 50 ohm resistors you can place two in series for each bank to make them 100 ohms.
 
  • Like
Reactions: Shadetree Mechanic
Played with this some more over the weekend and something is definitely out of whack. I reconnected the driver, all back to “stock” and it’s just not producing decent power. I can drive the living snot out of it and show 350 pep, but there’s RF everywhere and it’s drawing a boatload of current for that kind of power...pretty well the same results I had with the driver bypassed.

Somethings way out of tune. My Daiwa meter really doesn’t like it.

Not sure what I’ll do now...can’t decide between replacing all the caps, and continuing to work on the input, or tearing it down for the transistors and rebuilding on a new board...nice to have the Toshibas, but I highly doubt they’re matched judging by the resistor on the combiner that’s dang near the size of a Vienna sausage.
 
Ok...the follow up to that is, am I good at ~.7v for bias voltage for a pair of 2879s, or do I really need to be measuring current?

I used a diode and 50ohm sandbar per pair, with bypass caps from transformer to ground and across the diode.

.7 volts ended up being 400ma on a 2 pill fatboy that I biased.
 
  • Like
Reactions: 2NC995

dxChat
Help Users
  • No one is chatting at the moment.
  • dxBot:
    Tucker442 has left the room.
  • @ BJ radionut:
    LIVE 10:00 AM EST :cool:
  • @ Charles Edwards:
    I'm looking for factory settings 1 through 59 for a AT 5555 n2 or AT500 M2 I only wrote down half the values feel like a idiot I need help will be appreciated