[AMRadio] 6U8 Subststute and the 20a

Jim Candela jcandela at prodigy.net
Mon Nov 12 12:50:39 EST 2007

----- Original Message ----
From: jeremy-ca <km1h at jeremy.mv.com>

How did you handle the 8877 conversion as far as mounting and cooling?
DTR-2000 had the tube horizontal and blowing all the heat right on the
circuit. That caused a bit of tuning drift as the fixed caps for 80/160
right in the path; more so when the PS was upgraded to 3500V.

Ive a pair of MLA-2500's here being converted to 6M for customers, 
thankfully the 8875's are full power. But Ive been looking for tube
ideas that get rid of the heat better.


Reply From Jim, WD5JKO,


I have a couple of pictures at the following link, and I've pasted  some text below that I've already written concerning the modification. 



>From Dentron Yahoo group post that I entered:

Scott, I believe that one of the 8875 tube failures are the result of

a parasitic. I bet it is usually localized to one socket too. The

issue has to do with how the sockets are paralleled, and the blue ALC

wire takes the tube input RF and then routes it near the plate

circuit. This definitely creates a tendency towards a parasitic. One

or more of the MLA-2500 Conversions address this Dentron design issue.

I converted my MLA-2500 to a single 8877. I just added two pictures

under the files section labeled, "another MLA-2500 conversion". I had

to deal with the blue wire issue as well since the 8877 would take off

on peaks. Actually I disabled the ALC function, and used the

transformer ALC winding to "boost" the HV B+ (details upon request).

The 8877 equipped MLA-2500 can peak 2kw PEP out on SSB(for 1-2

milliseconds until the B+ sags), or do 400 watts out on AM 100%

modulated (1600 w PEP) with 20 watts carrier drive (80w pep). I did

need to add an additional fan on the top cover to help exhaust all

that heat from AM operation. The stock fan is blowing in from behind

but is spaced back by the Hammond box (containing the socket). The

Silastic RTV chimney was bought from the Alph folks.


Reread what I said. I stated that the amp would do ssb at 2kw pep

out for 1-2 milliseconds or until the B+ sags. This is completely

different from key down CW tune up for maximum power. The MLA-2500

power supply cannot maintain the B+ over 2250v at > 500ma CCS. My

power supply modification boosts the B+ by ~ 150v under load. Even so,

brief tune ups approaching 1 ampere result in the B+ sagging below 2

KV. As a driver during my testing I was limited to no more than 55

watts carrier (and 65w pep ssb), so I didn't worry about excessive

grid current from over zealous drive beyond that. With my drive

limited to 55 watts the maximum CW output is limited to 900 watts (B+

sags a lot). If I had a real power supply then things could would be


So to answer your question, I can see ssb at 2kw PEP on an

oscilloscope briefly (1-2 ms) from the idle B+ of about 2600vdc with

about 65 watts pep drive. The scope shows uncompressed voice peaks

over 316v peak (or 632v peak to peak) into a 50 ohm load. The only

fire so far is an occasional tuning capacitor flash over. I did have

to play some with those fixed loading capacitors that were used

improperly and beyond ratings. I replaced some of them with 2 or 3

smaller values so as to lower the RF current through each capacitor.

Using bypass capacitors in a hi-Q tuned circuit is not always a good

thing. As they heat up the capacitance changes, and therefore the

tuning changes. Not good. AM operation at 400 watts carrier out

revealed the loading problem.

Thanks for the feedback,



More information about the AMRadio mailing list

This page last updated 22 Feb 2018.