Sinus Rig Repair

[sinus trouble]

Hello Necks
I built this rig a few years ago and its been running great, even in quite harsh weather conditions!
However the trimmers on the driver board began causing problems and shorted causing the DC regulator to shut down! Luckily the 2n4427 seems to have survived!
Anyways ive changed out the cheap yellow trimmers for Teflon types! (As can be seen in the pic)

20170312_230528.png

But that's not all! I was pushing my luck with the 7815 regulators, The DC input is around 40v so I need to re think the PSU???
I could change the transformer for a lower voltage and lose a small amount of power for reliability?
Or maybe a limiting resistor to each regulator?
What do you think Necks?? Any ideas would be appreciated!

20170312_230441.png

[Albert H]

The simplest retro-fit solution would be to build a zener and power transistor regulator to give you (say) 28V into the regulators. That way you wouldn't be stretching them. Depending on the current you need, you could choose any number of medium power transistors. Remember to RF bypass each junction in the pre-regulator! Use a couple of (say) 12V zeners in series as the reference.

[thewisepranker]

The LM317 can handle 40 V across Vin to Vo. Or are you having thermal problems? A single pre-regulator will have to handle the thermal load of three times roughly half the current thermal burden of each regulator, which is a worse scenario to be in. Assuming you've got +15, +12 and +5 V rails.

[sinus trouble]

Many thanks ppl!
The regulator on the far left is a 7824 and runs the 24v fan
The center regulator is a 7815 and is just for powering any external STL receiver if connected
Finally another 7815 on the far right runs the driver/PLL board

As Mr WP touched on, the 7824 I believe also has a max rating of 40v and has performed great!
I haven't noticed any sign of thermal issues? most likely due to the low consumption?
All evidence seems to be showing that the 7815s were already on the brink of destruction (over max voltage)
and the trimmer failures pushed them over the edge! lol

Mr Albert! Maybe I could use a similar idea by using a power transistor shunt style across the 7824 which in turn could supply the fan and both 7815s? similar to this?

LM7824-circuits.png

[thewisepranker]

The 78 series regulator rating is relative to GND, whereas the LM317 is Vin - Vo. Even with a 40 V capable 7824, 40 V is the absolute maximum rating, not the typical rating. Your unregulated DC voltage is not going to be an integer and is more than likely quite a bit over 40, hence the blue smoke will escape and then it's knackered as you'll never get all the smoke back in again.

Use an LM317 to feed the two 15 V regulators if the power dissipation is sufficiently low. All you need is two additional resistors to set the voltage of the LM317.

[radio-berlin]

Me personally wouldn't do any of the above. I'd personally wind 10metres of wire around the toroidal, ptfe equipment hookup wire is fine (standard hookup wire from craplin works fine), this will give you around 18volts ac (guessing you have 36v transformer), sling in a bridge rectifier and smoothing cap and you have a completely isolated supply capable of a couple of amps or so

[Analyser]

An alternative would be to use a switching buck regulator. You can find them all over the place using LM2575 or LM2576 made by ST.
The only thing you have to do is provide some good filtering or even stick a 7815 on the output to get rid of the switching artefacts that these type of things produce.

Decisions decisions!!

[sinus trouble]

Decisions indeed!
Many thanks to you all for your contributions!
Im gonna give the LM317s a try as 'Plan A' lol im sure I have some of those lying around here somewhere?
If that goes horribly wrong? atleast now ive got a Plan B an C! lol

[radium98]

Why not taking the input of the reg7815 from the output of the 7824 and follow the 7815 by a TIP35C work nice to me very old good technologie still work nowadays better .
as it seems to me a veronica NRG vco with pira PLL how di you connect the pll to it ,you tune the vco and then the pll ? because it is not a wide band .

[sinus trouble]

Thankyou Mr Radium
Yes its a nice little Pll exciter! I have built many and although not perfect? they work well

20140501_220700.png

The RF to the Pll comes from the input of the first yellow trimmer, then control voltage is applied to the varicap! The DIP switch is set before power up, then tuned up like a simple VFO!

The fundamental problem with my rig lies with the PA! It is based on the BLF177 which is a 48v device, to get any significant gain I need to pump as much voltage in as possible, especially with only 1watt drive!
This in turn put serious stress on the regulators which have an absolute max input of 40v! Whilst they worked fine for a while? They eventually 'Died' lol
Mr Wise pranker suggested using LM317s which are 'floating' regulators and if I have understood correctly? Means that an LM317 operating at an output voltage of 24v will handle a Max input of 64v?

I could be completly wrong though???

[Albert H]

Don't waste time and money on the 317. It's actually rated at 37V maximum input.

All you need is a 30V zener (or a couple of 15V ones in series), a resistor to put a bit of current through the zeners (2k2, 3k3 or thereabouts) and an NPN Darlington power transistor (I use BD679). Look up "series regulator" on Google!

[thewisepranker]

No it isn't. It's rated at 37 V maximum output.

On Semiconductor:
"The LM317 is an adjustable 3−terminal positive voltage regulator capable of supplying in excess of 1.5 A over an output voltage range of 1.2 V to 37 V. "
Maximum ratings:
Input−Output Voltage Differential VI−VO −0.3 to 40 Vdc

Texas:
"The LM117 and LM317-N series of adjustable 3-pin positive voltage regulators are capable of supplying in excess of 1.5 A over a 1.25-V to 37-V output range and a wide temperature range"
Absolute maximum ratings:
Input-output voltage differential −0.3 40 V

STMicro:
" They are designed to supply more than 1.5 A of load current with an output voltage adjustable over a 1.2 to 37 V range."
Maximum ratings:
VI - VO Input-reference differential voltage 40 V

[Shedbuilt]

Indeed. The LM317 doesn't have a grounded pin, so it doesn't have the full supply voltage across the pins of the device.
The "Adj" pin is 1.25V below Vout, so the largest voltage across any two pins, is the differential voltage (vout - vin).
It can be done, but a linear regulator, wouldn't be my personal preference. I don't like having that kind of voltage drop across a linear regulator; especially with significant current draw, because it is so inefficient.
Yes. the LM317 can supply more than 1.5A, and with a differential voltage of up to 40V.
If you drop 40V across the device, however, and draw 1.5A, you have 60W of heat to get rid of, and you're wasting that portion of the PSU capacity too.
Even at below an amp draw, if you're dropping tens of volts, you have a lot of unnecessary heat. You can split the drop across multiple devices (like Albert's zener / dropper transistor solution), which eases the load on individual devices, as well as dropping the input voltage, but you're still wasting the same (or very slightly more) power, and you still have to get rid of that, as heat.
I've seen 78 series regulators go into thermal shutdown, where this hasn't been properly considered - without exceeding any of the device maximum ratings (except temperature).
With 50v output devices especially, I always preferred - where possible, to run the 12v and 15v regulators from a lower voltage tap on the transformer (in some cases using a centre tapped transformer), and RF Head's suggestion will give you that lower voltage tap. That, or the DC-DC converter suggested by Analyser (with consideration - as he said, to eliminating the noise), would be my personal preferences.

[radium98]

thank you BUT my last suggestion to be is to use a switch mode power supply rated between 44-48v 10amps and not use torric transfos.youre rig will be light weight and less consomption in current.no hum no heat but i dont like switch mode ps because in my country dont have earth where at most time at start up the fuse blown due the electricity phase problem
OFF topic who can help me witjh a little STL link no so complicated as i mentioned in a thread before please mates

[sinus trouble]

Thankyou Mr Shed for your advice!
Its so difficult now? I don't wanna dismiss any suggestions, they are all excellent solutions to my dilemma!
I need to decide based on a compromise between complexity, reliability and the space I have to house such a unit?
Mr Analysers method seems to be very efficient? Theoretically I am familiar with DC to DC converters? However putting it into practice is another matter, I could be biting off more than I can chew! lol

Mr Berlins method is also very efficient and not so tricky to implement! The transformer has two independent 0-30v secondary windings @5amp each! So I could modify one winding to achieve this? But a breach of the insulation could be an issue?

Mr Albert and Mr Prankers are similar in efficiency! As stated high current applications could cause problems? After checking the load? the consumption is quite small? 86mA @24volt for the fan and around 160mA @15volt for the PLL/Driver?

[sinus trouble]

Hello Mr Radium
The SMPS is an excellent option and if it was a New build? I would defo consider it! This rig has done well over the years and will soon be retiring lol
In reply to your question regarding STLs? I suspect no one answered because it is Vague to say the least? Law and frequency selection will affect depending on your country of origin?
Building a stable STL transmitter from scratch is no easy task! And the receiver gets even more complicated!
If you are looking for a short range simple fix? All I can suggest is audio RF modules?

watxrx03-high-end-ism-modules_ace6c8d90426673c2c16747a45528ae1.jpg

[thewisepranker]

NPN Darlington power transistor (I use BD679)

10 °C/W thermal resistance from junction to case with the SOT-32 package... TIP122?

I've seen 78 series regulators go into thermal shutdown, where this hasn't been properly considered - without exceeding any of the device maximum ratings (except temperature).

This isn't a phenomenon unique to 78 series regulators. It's good that they have the feature, not that it's a good idea to rely upon it.

If you drop 40V across the device, however, and draw 1.5A, you have 60W of heat to get rid of

...but we're not. We are dropping perhaps 10 V at not even 300 mA, given the post by Sinus stating current requirements.

With 50v output devices especially, I always preferred - where possible, to run the 12v and 15v regulators from a lower voltage tap on the transformer (in some cases using a centre tapped transformer), and RF Head's suggestion will give you that lower voltage tap. That, or the DC-DC converter suggested by Analyser (with consideration - as he said, to eliminating the noise), would be my personal preferences.

In an ideal world, you'd have another secondary wound at a lower voltage to run the lower voltage stuff, but since it's all boxed up and ready to go and given the fairly low current, the regulator (or power Darlington) will only be dissipating 2.5 Watts, which is well within the realm of just bolting it to the case.

It's not like we need to save power - it's a transmitter drawing hundreds of Watts.

I would just fit a regulator (whether you make one yourself using Zeners and power Darlingtons, or a pre-packaged device) upstream of your three existing regulators. It's not efficient but it doesn't matter. You will need to drill one hole and can solder the components directly to the legs of whatever TO-220 packaged device you end up using without needing a PCB or Veroboard.
Switch-mode power supplies will only cause you more problems than they will solve (you only have one problem to solve as it stands) and so will the great expense of either replacing your current transformer with one that has at least one other secondary, or adding a second separate transformer, bridge rectifier, smoothing capacitor(s)...

[Shedbuilt]

I've seen 78 series regulators go into thermal shutdown, where this hasn't been properly considered - without exceeding any of the device maximum ratings (except temperature).

This isn't a phenomenon unique to 78 series regulators. It's good that they have the feature, not that it's a good idea to rely upon it.

Agreed on all counts. Just making the point that heat needs to be considered, and isn't always.

If you drop 40V across the device, however, and draw 1.5A, you have 60W of heat to get rid of

...but we're not. We are dropping perhaps 10 V at not even 300 mA, given the post by Sinus stating current requirements.

Agreed in general, but I was talking there about the maximums previously referred, for the LM317. For Sinus' case, my expectations were 40V input, with 12v or 15v out (so 25 to 28v drop) - unless it was just used as a pre-regulator. I also expected the exciter consumption to be a bit more. I forgot which Veronica / NRG design it was based on (basically only two RF stages; the oscillator / doubler drives the 2N4427 directly).

With 50v output devices especially, I always preferred - where possible, to run the 12v and 15v regulators from a lower voltage tap on the transformer (in some cases using a centre tapped transformer), and RF Head's suggestion will give you that lower voltage tap. That, or the DC-DC converter suggested by Analyser (with consideration - as he said, to eliminating the noise), would be my personal preferences.

In an ideal world, you'd have another secondary wound at a lower voltage to run the lower voltage stuff, but since it's all boxed up and ready to go and given the fairly low current, the regulator (or power Darlington) will only be dissipating 2.5 Watts, which is well within the realm of just bolting it to the case.

It's not like we need to save power - it's a transmitter drawing hundreds of Watts.

I would just fit a regulator (whether you make one yourself using Zeners and power Darlingtons, or a pre-packaged device) upstream of your three existing regulators. It's not efficient but it doesn't matter. You will need to drill one hole and can solder the components directly to the legs of whatever TO-220 packaged device you end up using without needing a PCB or Veroboard.
Switch-mode power supplies will only cause you more problems than they will solve (you only have one problem to solve as it stands) and so will the great expense of either replacing your current transformer with one that has at least one other secondary, or adding a second separate transformer, bridge rectifier, smoothing capacitor(s)...

Agreed again. Another possibility, with a three legged / fixed regulator, is to put a Zener diode in series with (what's normally) the ground leg. This will raise the output voltage by the value of the Zener diode, and raises the maximum input voltage by the same. It does however (obviously) mean that the device needs to be insulated from the case. So you could use a three legged (eg 78 series) regulator, treated in this way, as a pre-regulator; for subsequent three legged regulators wired in the normal way.

[Analyser]

Boom!
http://enigma-shop.com/index.php?option ... tor-module

[radium98]

Thanks all