Alternative transistors to 2n4427 @ 2cs1947 (used on NRG PLL PRO III - PRO IV boards)

[rigmo]

2n3725a silicon npn transistor to39 national-semi

[radionortheast]

Its been along time since I had nrg transmitter, I do remember the output transistors on nrg/veronica’s would bust easyly. It might be worth trying one of these 2.5w rf amplifier modules, imagine it would be quite easy to route the rf in, using some thin coax, maybe feed the output back thought the filtering on the transmitter, or make your own, suppose it might work and provide more robust output, but it won’t go to 4w.

[rigmo]

we also have bad copies, but the best of the measured ones is the mysterious 2sc830

[reverend]

Are you judging which is 'best' by looking at the hFE? That usually only tells you the DC gain of the transistor, nothing about how good it is at producing RF or what gain it might have. If you could measure the 'fT' (the transition frequency), that would at least tell you whether any of the transistors are likely to work at RF frequencies.

[rigmo]

Are you judging which is 'best' by looking at the hFE? That usually only tells you the DC gain of the transistor, nothing about how good it is at producing RF or what gain it might have. If you could measure the 'fT' (the transition frequency), that would at least tell you whether any of the transistors are likely to work at RF frequencies.

That we can see quality swing.. you are right all about.

[rigmo]

For sale heatsink TO-5 TO-99 TO-39. IF you need better temperature dissipation than most other coolers. the minimum conditions to meet this are the presence of air currents with some small fans or only air currents from the entire system.
aviable Length 8mm and 25mm

[Albert H]

I don't understand Radionortheast's comment about the NRG stuff having easily damaged transistors - the 2SC1947 in the PLL Pro III board will withstand an open or shorted output for quite a while (as long as it has its heatsink on!), and the BLW60 used in the 40W PA was virtually unburstable! The 40W rigs had just the right fuse feeding the PA, which would blow if it was run into a bad mismatch for long. I remember seeing several returned rigs with "alternative fuses" on the PA boards!

[Shedbuilt]

I remember seeing several returned rigs with "alternative fuses" on the PA boards!

Good old Bacofuse

[LeeCavanagh]

I’ve got an issue with 2 separate. Nrg boards. The Pro III definitely gave me 4watts up until recently. But now it’s just under 2 watts. I have the same issue with the pro IV the larger heat sink on both still get very hot to touch (as I would expect) but the power doesn’t seem stable, it’s noticeable on the IV as the led power light slowly dims as power decreases (turning it off till it cools down appears to make it go back to 2watts -I find this strange but someone may have seen this issue any times ? Any ideas? Does it need the transistor replacing even though it’s giving power? Or could it be the smaller pre-amp as neither board appears to give any heat on the small heat sink (I know this never gets very hot. But I expect a little at least)

I’m also shocked I’ve had the same issue with both , never had this problem since they were designed, then I have 2 do the same thing in the same week. I even changed my dummy load just in case it was that.

[Albert H]

Have you ever put them into a bad mismatch? The two big resistors in the final stage can sometimes "cook" and rise in value, reducing the supply voltage, reducing the output power. However, it's probably more likely that you've got failing 2SC1947s - if you've run into a mismatch for any long period of time, some of the emitter bonds inside the transistor will have failed, giving increased emitter resistance, and therefore reducing the transistor gain. I never really liked those tubular heatsinks on the '1947s, and was in the process of designing a new output stage for the Pro III, firstly using a 2SC1971, and then later, using one of the Mitsubishi FETs (RD06). This necessitated board layout changes, and in the case of the FET option, we would have had to add a bias circuit, but it would have made the rigs cheaper and more robust....

[LeeCavanagh]

Have you ever put them into a bad mismatch?

Possibly (h hmm cough cough), can’t say for sure.
Didn’t you say

the 2SC1947 in the PLL Pro III board will withstand an open or shorted output for quite a while

How long is a while?
I guess it could explain why both have ended up the same way, but would that result in the power coming back after cooling for a bit ? I tried adding another heat sink on top with some silicone compound between, same thing tho , but my first thought was transistor. (It does explain why my separate PA was loosing power overtime . As the exciter was getting lower in power over time, possibly as low as not even being able to trigger the pa transistor… that aside tho)

I guess I could try replacing them both in one board and if it works do the same on the other, is that smaller transistor unlikely to need replacing if that 1947 is heating up ?

I feel I really should know this
which resistor should I check with my multimeter ?
Thanks Albert

[yellowbeard]

R68 and R69 - they are the two fat 2 Watt ones, they should be 33R. Hope I got that right - I had to go root in the room of busted shit to find that out...

[LeeCavanagh]

Ok, thanks. I have one of those rooms too

[radium98]

Albert is write , this pictures are inside a genuine c1947 , the powerful transistor .

[XXL]

Does the NRG PLL PRO 3 go down to band 1 frequencies ?

[LeeCavanagh]

Thanks for the visuals.

[Albert H]

Does the NRG PLL PRO 3 go down to band 1 frequencies ?

Not as it's normally built. However, if you wind the oscillator coils with 3 turns on each, you can get down to about 63MHz. If you wind them with four turns, the VCO output will be around 28 - 55 MHz. You also have to increase the trimmer capacitor (though if you can accept a limited range, you can just pad it with a capacitor in parallel). Bear in mind that the driver and output stages will also need mre turns on the coils and increased capacitor values.

Way back, Steven Moss did make some Band 1 link rigs using the Pro III boards, but I don't think I have details of all the changes he made to move them down in frequency. He paired them up with some of my link receivers for RSL stations. My link receivers used the UV616 TV front end which covered every frequency from the bottom of Band 1 right up to around 1 GHz in several bands. The only problems with the front end were that the IF was 35MHz and the IF bandwidth was huge (to let colour telly bandwidth signals through). My fixes were to mix the 35MHz IF with a crystal to give a 10.7MHz second IF, then use a conventional CA3089 at 10.7MHz, and I also added a filter stage to the 35MHz IF to tighten the received bandwidth. The receivers were cheap, fairly simple and easily configured for almost any conceivable link frequency!

[LeeCavanagh]

Will the 2N3553 give the same 4-5w ? Or potentially more (as I notice it’s dated 7w)?

I tried two diffeeent 4427 in my board and both gave me less than 0.5w, maybe I’m unlucky (one was genuine but was taken out of another board which was working when it was taken out I believe. The second looks like a brand new but Japanese copy so that probably gave up before it got started - still heats up very hot (this always surprises me, if it’s only giving out 0.5w then why does it still heat up just like it’s giving out 4w…. But anyway , do all the alternative s listed on the first post all give out the same power in the NRG configuration?

[Albert H]

The 7 Watts in the datasheet for the 2N3553 is maximum dissipation, not power output! In reality, you'll get just slightly less than half of the dissipation as RF - typically around 3 Watts at best. The '3553 was a pretty horrible transistor - it was OK at HF, but ran out of gain at more than about 50MHz (unless you were very lucky). At one time, it was just about the only device available with that kind of specification, so many of us were forced to press it into service. Back in the early 80s, I did a PA with a pair of them driving into a 2N5643 for about 35 Watts out. It probably made much more heat than RF, but it was reasonably clean and a couple of stations used them for a while.

I really don't understand the obsession with all these ancient transistors. These days you have things like the RD01MUS1 which will give you an easy watt when driven with a simple BFR96 oscillator with an 8V supply.

I have a little ~1 Watt exciter prototype with an LMX2306 PLL and a PIC, using a couple of BFR96s in a self-doubling oscillator configuration a bit like the NRG circuit (but with amplitude-levelling), and one of the little Mitsubishi FETs for the output. The power output can be set anywhere between 500mW and 1.5W (to suit a range of PAs), and it's completely broadband. Frequency setting has to be done by flashing the PIC at present, but I plan to do a version with switches to select frequency. Unfortunately, adding the switches will almost double the board area!

Incidentally, there's also a Band V version of the same circuit, using etched Lecher Line coils on the PCB, using almost exactly the same components. This will make a great little 1 Watt stereo link rig!

[LeeCavanagh]

Thanks al,

My pa needs 4-6 watts input for about 500-650wi did consider just using the nrg at 1w then a 5w pa to go into the main amp (is that a bad idea?), probably not worth the extra effort just to squeeze out another watt (oh, I tried that idea of joining the two large pads on the underside of the III but it made no noticeable difference for me - maybe because of the transistor issue)

Will your prototype be shared on here? Or is it for your commercial ventures?
Also did your friend get any closer to finishing his new exciter with the 1971 output stage?

Of course, as the pro III and IV become thin on the ground then we need to look elsewhere, so your ideas are always of interest, and everyone on here (like rigmo) who are building tweaking etc, all great.

I found an old print out from NRG, for a 5w pa (which it says ‘should work but was never used in production’) you probably know the one I mean.

Something i noticed on one of the IV pro boards was the variable power resistor, the power would drop till tapping it with a screwdriver, solder was fine, I replaced the variable resistor , I did wonder if it’s not really rated for the job - clearly it failed somehow or has someone has been rough with it.