Along the same lines as I mentioned the other day, QRO Technology made a blunder on the HF 2500 amplifier. ?I found this problem when a local ham called me and told me that his HF 2500 would restart the initial warm up timer sequence every time he switched the fan speed switch. ?I reiterated what he said and he confirmed I didn’t hear incorrectly. ?i?said to him that is next to impossible and he said it does it every single time without fail. ?I kinda thought it was nuts but I said bring it over and I’ll look at it. ?So he came by and I hooked it up on the test bench. ?Turned on the amp with the fan speed switch in low speed. ?Amp started up and went through the warmup cycle. ?I then switched the fan speed to high and damn if the warm up cycle started up again. ?Well I saw it happened and I was a believer. ?I told him he would have to leave the amp while I tried to find out why that was happening. ?I started out ?by trying to wrap my head around the problem by thinking to myself.how the two circuits might be intertwined electrically. ?No matter how I?tried to do it there simply was no relationship between the two circuits. ?Now I’ve seen the timer circuits in these amps go crazy with leaky tantulum caps and a bad transistor and on some that used old carbon resistors. ?So I just went in and replaced all three. ?Afterwards it did the same thing. ?That’s when I pulled the schematic and mulled over it. ?I particularly concentrated on the fan speed circuit. ?Drawing on my many years experience as an electrician I know full well the difference on switching resistive loads and inductive loads. ?Subconsciously we probably all know when we switch on a gang of old fluorescent lights. ?Remember that familiar click ?In the switch box? ?Well when I saw the way they wired the speed switching of the fan I instantly knew what the problem was. ?So I went and rewired the fan speed switching circuit, end of problem.
On Sunday, January 5, 2025, 1:02 AM, John Sparkes via groups.io <vk6jx1@...> wrote:
Hi fellas
Sounds like you guys learned in your hobby what I used to teach my junior electrical engineers. Of course, they all worried about the hot wire. I told them in order of importance, it’s earth, then neutral, with the hot wire way down the list. The green (E) and black one (B-) will kill you or land you in prison (as the designer). Same applies to amps !
John
On 5 Jan 2025, at 2:03?am, Jim VE7RF via groups.io <jim.thom@...> wrote:
?
Lou can chime in on this one.? On? the..... "big 3"? ? ameritron amps....... they don't use a resistor at all.? On RX it's wide open !? ?I got into an argument with W8JI about this years ago.? ?IMO, it's a stupid idea...... and 'saves' all of a few pennies at best.? ?While it ( a wide open across the relay contacts)? might? just work on the ameritron amps...it sure as hell doesn't work on that single and also dual GS35B? 6m amp I designed for a buddy in W7 land years ago.? ?I contend that with the B-? in the cathode open on RX........? the cathode will? try and assume partial to full B+.? ?The handbooks say the same.?
On those GS35B amps, the builder used red millens for the B+...and black millens for the B-.? ?One day, while putting it back together...( HV supply in a separate box) the black millen was not making contact.? ?With fil lit, then the B+ on next, all hell breaks loose...with an open B-.? ?The cathode indeed assumes B+ potential.? ?Those amps had aprx 4500 vdc no load on em.
I had the? exact same thing happen on my hb 4-1000 amps.? ? HV was in it's own box...... and separate? connections for the B+ ( millen)......and similar for the? B-.? ? One day, in super haste, zero patience....and only being 20 yrs old back then,? I wired it all back up....and proceeded to get in on the contest that night.? ? Like an idiot,? I completely forgot to hook up the B-? at the amp end.? ?All hell broke loose... with loud snapping and crackling going on.? ?Hooked up the B- and al was well.?
Next time I had it apart, the .01uf @ 1 kv disc caps used on the cold end of the fil choke were fried... with huge black smudge marks down the side of em.? ?With no B- hooked up, the cathode assumed all of the? B+....and arced across the bypass caps.? ? ?I was lucky it didn't arc across the coupling caps... ( between cathode and tuned inputs)..and start rfying the caps on the tuned input.?
These days, I take all that B-? wiring seriously.? ?The B-? from the supply goes to the negative side of the plate current meter.? Positive terminal of plate current meter is bonded to the positive terminal of grid current meter.? At that junction point,? it goes off to the bias cutoff resistor ( 2 x 100k @ 3 watt MOF's in parallel....for redundancy)...then through a mess of series bias diodes, then to the? CT of the fil xfmr.? ?
IF the plate current meter ever went open,? you have just opened up the B-.....and again, all hell breaks loose.? ?So in goes safety diodes across the plate current meter ( in addition to safety diodes wired directly between B- of? HV supply....and the HV supply chassis...and? also between? B-? and chassis? in the RF deck. ).?
On a similar note, drake used this stupid scheme? for cutoff bias? in the L4B / L7 amps.? ?They applied +120 vdc (+90 vdc when in? the lower 1900? voltage CW position) to the? CT of the fil xfmr...on RX.? On TX the center contact removed the +120/+90 vdc.... and instead? the CT? went off to the metering etc.?
Problem is...when relay contacts? in mid air? ( rich measures called that....'no mans land')? the CT has gone open.? ? ?And again, the cathode try's for a split second, to assume? full B+ potential.? ?Their scheme works.... but after years of use, the center contacts of the oem 3PDT TR relay were fried...and I mean burnt so bad.... it would not pull normal idle current.? With amp keyed, the CT was open.? ( The 100w input +? kw output contacts? were fine).?
?
?The easy fix of course was to toss the +120 vdc bias junk..... and install a 100k / 3 watt resistor in the CT of the fil xfmr.... an re-wire the contacts of the center pole of tr relay so resistor is inline all the time...and shunted on TX.?
I also added 10 x 1n5408's for bias....and also grnded the grids etc, etc.?
