Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise


It is currently May Mon 21, 2018 7:47 pm


All times are UTC [ DST ]





Post New Topic Post Reply  [ 80 posts ]  Go to page Previous  1, 2, 3, 4
Author Message
 Post subject: Re: Collins R-390 solid state regulator (problem)
PostPosted: Jan Wed 31, 2018 10:50 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
Well I got the regulator rebuilt today.

When I tested it I at first got no B+ at all and while probing around in the audio chassis a slight shock told me that yes I had B+ to the audio chassis. More probing showed that I had the B+ connection to the regulator on the wrong terminal of a tube socket. Moving it to the right terminal got some B+, but not much and turning up the variac only caused the B+ fuse to blow.

An investigation shows I had the 1N4007 to ground on the output backwards :oops:

Corrected that and had proper B+.

I found by looking at the choke connections which was the B+ side and which fed the audio chassis so I installed that diode the correct way.

Will test further tonight. Bit leery about testing the two protection diodes by putting the receiver into standby as I am unsure if they will do the job, although the one across the choke should be all I need as it can handle 1000 volts at 1A and when the choke's field collapses it will forward bias the diode thusfore eliminating the high reverse voltage.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (problem)
PostPosted: Feb Thu 01, 2018 6:05 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
I tested it tonight and operated the standby switch several times and thought all was good. Operated the standby switch again and the regulator is toast just like what happened last time.

I have absolutely no clue what is causing the damage to the regulator.

I know it isn't the choke as there is a diode across it to prevent the back EMF and also there's a diode to ground on the regulator output that would protect the regulator from back EMF.

There is a 10uF cap after the choke which I upped to 100uF originally to fix a problem that I found was an issue with the local audio level control. Could that be causing the issue or is the choke's DC resistance high enough to where the inrush current wouldn't be high enough to do damage to the regulator?

With that I think I am done trying a solid state regulator of any kind unless someone can figure out what is damaging the regulator and can offer a solution.

Oldwirebender, that simulation you ran on the regulator circuit can you do it again with a choke of 4H 50mA max current 110 ohms resistance which has a reverse biased 1N4007 diode across it with a 100uF filter cap after the choke with a resistor across it to equal 40mA with the load resistance of the regulator increased to account for the 40mA going through the choke and a switch that will switch the regulator's load and the choke circuit in and out of circuit?

I want to see if there's a way with that simulation to see if it will identify the problem area as I'm not ordering any more parts until the problem is figured out.

If all else fails I will go back to tubes and hope mine last a long time and that some are available when they completely fail. Only thing that saved me when using the tubes is the fact that the original 18K power resistor got broken when I installed the mounting screw wrong and replaced it with a 25K power resistor with adjustable tap and found out that resistor sets the output voltage. When I found my B+ was 174Vdc (weak regulator tubes) I had adjusted that resistor for 180Vdc.

The regulator circuit was to do three things.

1. Eliminate two rare tubes.
2. Eliminate a lot of heat.
3. Eliminate the need for a fan.

1 it did

2 & 3 not so much as I still have plenty of heat (not nearly as much as the regulator tubes) and still need a fan.

Also the 100 ohm resistors in place of the 47 ohm resistors didn't do as much to drop the B+ as I thought they would.

I'd like to get the unregulated B+ down to around 250Vdc, but that will require a big honking resistor and will require drilling two holes on the power supply chassis to mount it plus it will generate heat which will increase the need for a fan.

I wonder if I removed the rectifier tube sockets would I have enough room for a choke and just make it a choke input power supply provided that would give me the right B+ under load?

Not sure what value of choke I'd need.

Also if I get B+ down to 250Vdc that will be much easier on the regulator and perhaps then I would not need a fan as less power will be dissipated in the regulator circuit.

This time I think I'll disassemble the regulator and see if I can identify the failed part or parts.

If it's just the LM-317 I can get those easy. Same with the 1N4148. The 1N4007, zener and MOSFET I would have to order.

I can say that switching between standby and AGC there was only around a 1mV difference in regulated B+.



EDIT:

Here's the schematic showing the choke and standby switch

Attachment:
R-390 regulator 5.jpg
R-390 regulator 5.jpg [ 106.82 KiB | Viewed 471 times ]


The failure analysis shows that both the LM-317 and 12N50E failed.

The short is between the in and out terminals of the LM-317 and source and drain terminals of the 12N50E.

Have no clue why they failed though.

I could maybe see the LM-317 failing if too high of a charge current is present with the 100uF cap, but there's no logical reason that I can see for the 12N50E to fail short source to drain.

What I would like to do is figure some way to drop the main B+ to 250-280 Vdc no load and have it remain there under load. That way the regulator will not have such a high input B+ voltage when in standby mode and when operating.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Fri 09, 2018 6:24 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
I'm thinking it might be better just to use an external DC supply capable of 180 Vdc at 200mA. I could plug it in using the 180 volt test point on the side of the chassis and connect ground to the same place the power cord is grounded to.

In addition to removing heat from the chassis it also reduces the load on the power transformer.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Sun 18, 2018 2:48 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2742
Location: Athens, Greece
Many years ago when I had an R-390 and no 6082 tubes were available anywhere, I just subbed it with a TV horizontal output transistor mounted on a large heatsink. No other components, as I recall.

The regulated voltage circuitry is already there, why complicate this? Several volts change in the output won't affect the stability of this radio.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Sun 18, 2018 4:38 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
I had tried a power MOSFET to replace the regulator tubes and it didn't exactly work and wound up damaging the MOSFET.

I had thought of a transistor, but someone had mentioned a MOSFET was the appropriate choice.

Forget why though. Might have been due to the higher gate resistance being more like a tube.

I do need to figure a way to get the main B+ to be lower though as that might be the actual problem.

The circuit was only simulated at 300Vdc input so maybe the higher input voltage which is closer to 400Vdc unloaded is causing a problem.

I initially tried several 20 volt 5 watt zeners with a previous regulator circuit to provide a constant drop in B+, but they ran hot enough to melt the solder so that idea was abandoned rather quickly.

Now if there was a way to make zeners work without them running hot enough to melt solder maybe that would work.

What I may do is try to run B+ at 300Vdc unloaded with the heaters at full voltage, replace the MOSFET and LM-317 then operate the standby switch several times and see if the regulator then fails.

I myself prefer keeping B+ voltage within spec since that's what the radio was designed to operate at and was the voltage I aligned it at.

The ultimate solution would be to have the transformer rewound for less HV AC output so that unloaded B+ is 300Vdc, but then only the listed circuit would work.

If all else fails I'll use an external regulated 180 Vdc supply.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Sun 18, 2018 4:52 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2742
Location: Athens, Greece
My mod seemed to work fine. I might have added a series dropping resistor to reduce some of the power dissipation, can't remember - that was over 30 years ago. I do remember that I built it on a good sized heatsink and wired it to an octal plug, which just plugged into one of the 6082 sockets.

Worth a try, a horizontal output transistor is cheap.

You could just go with the 6082s, they aren't that expensive (albeit not exactly in the 6AK5 category). In all honesty the regulation isn't going make much difference at all since the R390 is really stable to begin with; also, you're not dealing with field/generator power or some iffy temporary sigint camp like these radios were designed for. The 390A has NO regulation at all (apart from that stupid 3TF7 fil regulator for the BFO and PTO) and the stability far exceeds any of its contemporaries. It's not worth overthinking this.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Sun 18, 2018 7:36 pm 
New Member

Joined: Feb Fri 09, 2018 6:44 pm
Posts: 24
Here’s a thought have you ever used lm317 voltage regulators? Well the TL783 is a similar regulator capable of an input/output voltage differential of 125 volts at 700ma.

300v-180v+120v is close to the in/out voltage limit.
However I believe you could run 2 of these in series with the first one regulating from 300v to 240v feeding a second one regulating from 240v to 180v. At 200ma with 60 volts across each one they would each dissipate around 12 watts so would require a heat sink.

This would also do away with any high power resistors.

These can be found on eBay for about $1.00 each

http://www.ti.com/lit/ds/symlink/tl783.pdf


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Sun 18, 2018 11:49 pm 
New Member

Joined: Feb Fri 09, 2018 6:44 pm
Posts: 24
This is what I had in mind. I haven't built this circuit but can't see why it shouldn't work.

This is as close as I could get to 180v using standard resistor values but a resistor and trimmer could be used in place of R3 to set it to exactly 180v

Remember to avoid having a drop of more the 125v across either of the regulators.


Attachments:
Image1a.jpg
Image1a.jpg [ 30.65 KiB | Viewed 390 times ]
Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Mon 19, 2018 4:41 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
I like that circuit, but the issue there is that the unloaded B+ is around 400Vdc.

I'm sure if I used the rectifier tubes B+ would be lower.

What I need to do is use the power transformer I originally had for testing the initial solid state regulator circuit so I can vary the B+ independent of the main power transformer.

That way I can set the unloaded B+ to 300Vdc and then when the receiver is warmed up I can operate the standby switch several times and see if the regulator then fails.

I'm thinking that somehow some rating is being exceeded when switching from no load to load as the B+ does increase when unloaded.

That said given the ratings of the MOSFET I cannot figure out what is causing the source to drain short while leaving the gate unaffected.

With the protection that is already present plus the two diodes I added there should be no back EMF so the voltage rating of the MOSFET shouldn't be exceeded. The unloaded B+ voltage isn't too high for the MOSFET and I'm sure there isn't nearly enough current drawn at any time to exceed the current rating of the MOSFET, besides if the current was over 3/8A the B+ fuse would blow and it never has.

I could go with the 6082 tubes, but I would have to buy new ones plus I would be defeating one purpose of the solid state regulator which is removing some of the heat from the radio. Plus as 6082 tubes get more rare and expensive I'd wind up having to do this in the future anyways.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Mon 19, 2018 4:54 am 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 36509
Location: Livermore, CA
Using solid state will save heat from filament of 6082's. Solid state regulators still will need heat sinks.

6082's aren't that scarse. They are $12 here:

http://www.thetubestore.com/Tubes/6000- ... Types/6082

One disadvantage of solid state. If it fails shorted full voltage would be coming from the regulator. That shouldn't happen with 6082 tubes.

_________________
Norm


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Mon 19, 2018 5:30 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
Yes the regulator needs a heatsink, but there won't be as much heat especially if I get B+ low enough to where it still regulates properly under load.

Yes if the regulator does short full B+ at nearly 400 volts is on the output.

Need to figure out a way to solve that issue. Perhaps one of those gas discharge tubes used to protect certain things from over voltage. If one is available that works at 250 volts should the regulator fail the tube would fire shorting the output and blowing the 3/8A fuse.

Didn't know they were $12 each.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Mon 19, 2018 7:09 am 
New Member

Joined: Feb Fri 09, 2018 6:44 pm
Posts: 24
RE: the TL783 circuit a third stage could be add to drop the 400v to 300v (82 ohm and 19,6K ohm) This should look after the high unloaded voltage.

The TL783 has thermal shutdown protection and is current limited to 750ma but I don't know how it would fail open or shorted.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Mon 19, 2018 6:40 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
Now would it be possible to add the TL-783 to the current solid state regulator?

If so that might work to drop enough voltage, but I need to rebuild the regulator and see what the unregulated B+ is under load first before deciding how much of a drop I need.

I do have a regulated HV supply I am purchasing so if all else fails I may use that.

The way things are now I can easily reinstall the original regulator by connecting a few components and wires and plugging in three tubes.

What I would like to do is find another audio chassis and leave that original then permanently mod this chassis so that I have room for things such a better audio transformer that drives the audio filters and a better heatsink for the regulator circuit.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Mon 19, 2018 7:34 pm 
Member

Joined: Sep Tue 30, 2014 6:08 am
Posts: 2541
Location: The Old Dominion VA 23518
The only down side to using cascaded TL783 devices is that the case (the "tab" of the TO-220 case) is at output potential. As each regulator will have to be heatsinked (~55V drop, 200mA will mean 11 Watts of dissipation), each heatsink will be at a different potential. TI outlines on the datasheet a scenario using a high-Vce transistor to drop the first 100V or so, saving the need for a large cascade of regulators.

I used three TL783's to build a 45-160V/400mA power supply for TV servicing. Two TL783s did the voltage regulation, and the third did the variable current regulation. It worked well for our application - we needed a supply to substitute B+ in sets with regulator/SMPS problems - it saved us from situations where we fixed the regulator/power supply in the set, only to find a bad flyback, CRT, or other issue. I used a MJ413 transistor to drop the first 60V or so.

_________________
Brian
"Capacitor Cosmetologist since 1979"
USN Retired 1984-2006 (Avionics/Cal)


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Mon 19, 2018 8:42 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
Far as heatsinking goes I would do what I did with the current regulator arrangement which is to use a transistor insulator and also an insulator for the screws used to mount the two devices.

But would such an arrangement of the cascaded regulators meet the +/- 1Vdc spec of the original regulator.

I know that the radio doesn't necessarily need to meet spec as I don't use it for anything that requires ultimate stability, but I do feel that a device should meet spec regardless whenever possible if only so if someone else acquires the device they don't have to worry about it meeting spec.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Mon 19, 2018 9:32 pm 
Member

Joined: Sep Tue 30, 2014 6:08 am
Posts: 2541
Location: The Old Dominion VA 23518
Using an insulator or Thermalsil pad works well, you just have to plan for more heatsink area. Could be a problem in a tight space.

In my power supply, I attached directly to the TL783s, and added a quiet 60mm fan on the side for forced air cooling. The MJ413 is hanging off the back, mica insulated from the external heat sink.

About a year after I built my TL783 supply, National came out with a totally insulated LM317HVP device - but killed it within a few years. With a 60V input-output differential possible, it would've been ideal for your situation.
Attachment:
to220p.jpg
to220p.jpg [ 4.29 KiB | Viewed 356 times ]

_________________
Brian
"Capacitor Cosmetologist since 1979"
USN Retired 1984-2006 (Avionics/Cal)


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Mon 19, 2018 10:34 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
That would definitely have worked to drop some of the B+.

The only thing I can think of that caused the regulator circuit to fail is the nearly 400Vdc B+ input, but doesn't explain the failure of the MOSFET given that as far as I know it isn't at any time above its ratings.

This time when I fix the regulator I will order two each of the active devices and do a test with full B+ and use my scope to see if I can figure out the problem.

If the regulator fails again I'll use the second set and try a reduced input B+ to see if that fixes it.

If the regulator then doesn't fail I'll know what the problem is.

If the regulator does fail then I may use the regulated power supply I am looking at getting and install a pin plug on the positive so it can plug into the 180 volt test jack on the side of the audio chassis. I will then connect the ground to the chassis.

Now if I find the regulator doesn't blow when the receiver is switched off and on multiple times by the power bar I have it plugged into I can just bypass the standby switch position of the function switch at the audio chassis and be done.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Feb Wed 28, 2018 2:40 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
Concerning the MOSFET can I use any MOSFET with similar specs?


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Mar Mon 12, 2018 3:07 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
I've got the regulated power supply on its way.

According to the seller it needs a recap and I will also have to find a way to set the output to 180Vdc.

I will just disconnect the diodes and resistors in the power supply. The test jack on the audio chassis will be used for the 180Vdc input and a chassis screw on the rear will be ground.

If the chassis as ground causes any problems I will remove the variable resistor next to the 180 volt test jack and install another jack for ground.

I'm hoping the supply is small enough to fit in the speaker cabinet. If so I will just set the amp on top of it with only the receiver on top of the speaker cabinet.

I also then won't need the fan drawing air out of the receiver.


Top
 Profile  
 
 Post subject: Re: Collins R-390 solid state regulator (circuit problem)
PostPosted: Mar Wed 21, 2018 12:43 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18578
Location: Warner Robins, GA
I like the regulated power supply, but it produces a bit of heat and it also buzzes some because of the large transformer.

So I more than likely will not be using it for very long with the R-390.

I'm thinking about buying the Edcor XPWR146 transformer https://www.edcorusa.com/xpwr146

That will provide an unloaded voltage of 282.8Vdc.

I will then use the solid state regulator which should work better with a lower B+ voltage.

The 800mA regulated supply may find use elsewheres.


Top
 Profile  
 
Post New Topic Post Reply  [ 80 posts ]  Go to page Previous  1, 2, 3, 4

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 2 guests



Search for:
Jump to:  
























Privacy Policy :: Powered by phpBB