Pages: [1]   Go Down

Author Topic: Astron RS-35M repair  (Read 609 times)

WB8PFZ

  • Member
  • Posts: 624
Astron RS-35M repair
« on: September 13, 2022, 10:49:35 AM »

Just acquired a used supply at an estate sale. Had blown fuse. Replaced fuse and meter showed 15+ volts. Adjusting the voltage control did not change the voltage. Replaced the 723 IC and all is well. Hope this will help out someone down the road. Mike
Logged

KD6VXI

  • Member
  • Posts: 1022
Re: Astron RS-35M repair
« Reply #1 on: September 13, 2022, 11:59:37 AM »

The LM723 is fairly prone to blowing.  I couldn't tell you how many I've had to replace over the last 35 years.  Probably more than a dozen?

Be nice if someone made a replacement that was more robust, but since DIP is pretty much behind us, I doubt it will happen.  Just keep a couple spares for your ham career and you'll be set.


--Shane
WP2ASS / ex KD6VXI
Logged

WB8PFZ

  • Member
  • Posts: 624
Re: Astron RS-35M repair
« Reply #2 on: September 13, 2022, 04:15:09 PM »

Luckily I had a few spares. Nothing worse than waiting a week or so and that's not the problem :)
Logged

K8AXW

  • Posts: 7391
    • HomeURL
Re: Astron RS-35M repair
« Reply #3 on: September 14, 2022, 11:42:00 AM »

This is why one should always take the time to install a DIP socket if the 723 fails.  It will serve two purposes.  1- The 723 won't fail again. (Murphy's Law) 2 - IF the 723 does fail again you can simply swap it out.

(The early Astrons as well as Astron clones had the 723 soldered into the PCB. I don't know about the newer ones.)
Logged
A Pessimist is Never Disappointed!

W9IQ

  • Member
  • Posts: 8866
Re: Astron RS-35M repair
« Reply #4 on: September 14, 2022, 11:59:09 AM »

This is why one should always take the time to install a DIP socket if the 723 fails.  It will serve two purposes.  1- The 723 won't fail again. (Murphy's Law) 2 - IF the 723 does fail again you can simply swap it out.

(The early Astrons as well as Astron clones had the 723 soldered into the PCB. I don't know about the newer ones.)

It might be a better idea to figure out why the 723 is failing and correct the design or circuit. It isn't an inherently unreliable chip. Done properly, the socket will be less reliable than the 723.

- Glenn W9IQ
Logged
- Glenn W9IQ

God runs electromagnetics on Monday, Wednesday and Friday by the wave theory and the devil runs it on Tuesday, Thursday and Saturday by the Quantum theory.

WB8PFZ

  • Member
  • Posts: 624
Re: Astron RS-35M repair
« Reply #5 on: September 14, 2022, 01:38:45 PM »

This one had a socket.
Logged

WB8PFZ

  • Member
  • Posts: 624
Re: Astron RS-35M repair
« Reply #6 on: September 14, 2022, 01:40:44 PM »

Without knowing who owned it or the history it would be hard to figure out what caused the failure. It also had a blown line fuse. Luckily those are in sockets.
Logged

AI5BC

  • Posts: 456
    • HomeURL
Re: Astron RS-35M repair
« Reply #7 on: September 14, 2022, 07:54:58 PM »

Just acquired a used supply at an estate sale. Had blown fuse. Replaced fuse and meter showed 15+ volts. Adjusting the voltage control did not change the voltage. Replaced the 723 IC and all is well. Hope this will help out someone down the road. Mike
This is a well know and documented design flaw with any Astron PS. Very easy to fix if you know how along with several other mods needed to make up for the poor deign.

The problem is how R5, the control loop feedback potentiometer is configured. The wiper blade is isolated from the other two poles and connected directly to pin 4 the inverting input. As the unit ages the wiper contact point becomes dirty and pitted which results in momentary open cwhich can immediately destroy the voltage comparator in the 723 instantly if the crow-bar circuit or input fuse to operate quickly enough. Classic symptom cannot control output voltage and output voltage is too high at V+.

Most of you guys and boat anchor owners remember crappy radio and TV volume controls when you turn the volume control up or down hearing scratch hiss noise when turning. It is the wiper blade going open circuit momentarily. Now what will really frost Astron's clingers is when you know how simple it is to fix and leave you wondering why Astron is 50 years behind the times and would not spend 2-cents to fix it. Real simple short the wiper pin terminal to the pin connected to R6 on the positive rail. So when you run across one of the pits or bad spots on the pot, the voltage is pulled high on the 723 Inverting Input resulting in the output voltage going lower rather than spiking high and F-ing things up. 

Got an Astron blowing fuses or operating the Crow-Bar circuit for no apparent reason, now you know what to look for and how to fix it. Even better replace that POS. They do make a good space heater or boat anchor with a heavy chain though.
Logged

K8AXW

  • Posts: 7391
    • HomeURL
Re: Astron RS-35M repair
« Reply #8 on: September 16, 2022, 10:25:40 AM »

5BC:  Thank you for the R5 info.  It has found a place in my Astron Power Supply file.

However, I think your feelings/opinion of the Astron power supplies are a bit harsh and unjustified.  Of course your opinions are entitled.

The problems with Astrons are what we deal with here on eHam.  We seldom hear anything about the thousands that are in use throughout the world that perform year in-year out without a hiccup.   

The Astron circuit (I have no idea if that it is an original circuit designed by Astron) is emulated by several other companies including Kenwood and Pyramid.  Circuit variations are inevitable but they all use the same old 723 regulator chip.

I've never considered the "R5 problem" and the 723s I've replaced have gone on to serve for years without a problem.  I do know that the 723 is susceptible to lightning. That subject has been well documented as well as how to deal with that problem. 

As for the rest of the failures, who can say what takes out the 723?  I've replaced the 723s, pass transistor failures,(mostly) rectifiers and even a  fuse that didn't work because the operator replaced a blown fuse with one that was too short.

I don't work for Astron but hate to see them get a bum rap because of a few failures.
Logged
A Pessimist is Never Disappointed!

AI5BC

  • Posts: 456
    • HomeURL
Re: Astron RS-35M repair
« Reply #9 on: September 16, 2022, 09:12:22 PM »

I don't work for Astron but hate to see them get a bum rap because of a few failures.

If you are willing to listen, let me point out details you have never considered or wanted to know. Let's start with why no ASTRON is UL listed. It has exposed live electrical components and unprotected metering wiring. All those exposed pass transistors have unregulated DC voltage, 19 to 23 volts depending on load. The collectors of the transistors are exposed and connected directly to the output rectifier/capacitor bank. Something comes in contact with the aluminum heatsink and collector, and BANG, dead short to chassis ground.

The panel meters have unfused and output voltage on them. Astron connected the panel wiring directly to the output without any current limiting device like a 2-cent 100-ohm resistor. They place calibration pots on the back of the meters rather than on the circuit board to protect the meter wiring—no UL cigar for Astron.

Grounding is a severe issue posing both safety and operational problems. Aside from the QC issue I pointed out earlier with the AC equipment ground chassis termination, the bonding of the chassis to the negative DC output is dangerous, destructive, creates RFI, and corrupts all and any grounding and bonding efforts. No other DC power supply manufacturer would do that because they know it does not comply with electrical codes and practices. All electrical codes and practices require all electrical systems to be Isolated from each other in the event of an electrical fault. If there is a fault (short circuit or high voltage), the fault current flows in all interconnected systems and can cause extensive damage to both systems. It is not rocket science, simple parallel circuit laws.

Hams compound the problem when they fail to comply with electrical codes. Hams intentionally place themselves in a ground loop, and your Astron makes it a DC ground loop. You run the coax directly outside to an earth ground (ground rod) rather than the AC Service ground before entering the house. Here is what hams do;  ground rod outside the shack to bond coax shield > coax shield inside to radio chassis > radio chassis to DC negative > DC negative to Astron chassis > Aston chassis to AC green wire in the power plug, power plug ground to AC earth ground aka service ground. You just placed yourself between two earth grounds with a serial daisy chain. You do not have a ground; your radio equipment is used as wire to connect two ground rods.

As a result, you get lots of problems you cannot fix or use tons of toroids trying to fix. With the DC negative bonded to the chassis, you lost your electrical and galvanic isolation between AC and Dc power systems, completely defeating the purpose of the transformer. All the noise currents in your home AC equipment ground system flow in your DC ground system and vice versa. You get a second DC negative wire. The one you know about, and the other is out on your coax shield to the ground and loops back around to the AC Service ground. You will have DC flowing on your earth ground system. What happens when you force DC current on ground conductors in moist dirt? Can you say galvanic corrosion destroying your ground system? Comply with codes and get rid of the jumper inside the Astron, and all that goes away.

Now let's talk about circuit design or the lack of it. I already addressed the voltage pot; however, another little tweak goes with it. If you cut the circuit trace to pin 4, the Inverting input and insert a 1K-ohm 1/8-watt resistor enables the Astron to be used as a Float charger, meaning you can connect a battery directly to the output with a fuse. No expensive high current blocking diode is needed.

The reference voltage amplifier has absolutely no filtering or soft start-start circuitry. A simple RC filter of a 3.9K ohm resistor and 10 uf cap placed between pins 5 and 6 give you an excellent soft start circuit and filtered voltage reference.
The LM723 power input is not regulated, resulting in poor PSRR. The LM723 has two power inputs: the regulator and the driver circuit. The driver circuit does not require a regulated voltage, but the regulator does if you want a power supply with decent PSRR. A proper design used a 78xx to supply power to pin 12 V+.

Current limiting and sensing are poor at best. Astron senses current via one of the pass transistor ballast resistors rather than a sense resistor on the output measuring the actual current. Inaccurate at best. Due to cutting corners and not using a series current sense resistor, Astron uses a resistor/diode network to simulate a current sense resistor to limit and control current, resulting in sloppy, inaccurate control. Compounding the issue is the 723 itself. If one looks at the Junction Temp vs. Current Limit, one instantly knows the 723 current limiter accuracy is poor.

RFI filtering, what RFI filtering? Astros lose regulation when used in solid VHF and UF environments like a collocated repeater site. Astron tried to address the problem 20 years ago by hanging an electrolytic cap across the output terminals. Any circuit designer or engineer with common sense knows that is useless because the ESR and inductance of the wire lead on the capacitor completely defeat the purpose.

The ICAS ratings are fiction. If you run the thermal analysis on the heatsinks clearly shows the rating is pure fiction. They get away with it because they know most hams do not push the limits. However, battery users, like repeater users, understand better. The first time your site goes on battery or places a discharged battery on an Astron without seriously reducing the current limit o 50% of the rating, it will require replacing the past elements. About the only thing most can do is run the output voltage as high as possible to minimize the thermal stress around 14.2 to 14.4 volts. Alternatively, if you know what you are doing is, add a 6/12 Vac buck transformer on the input to knock the transformer output voltage a couple of volts.

The bottom line is that the LM723 is an outdated late 1960s technology with high-side voltage control. It is inefficient, and when compounded by Astrons' poor design and QC, it is a poor choice when there are modern designs for less money carrying a warranty, UL, and FCC compliance.

For more reading, there is a website with a dedicated section on all the problems, fixes, tips, and tricks, along with every Astron schematic at Repeater Builder.
Logged

W9WQA

  • Member
  • Posts: 990
Re: Astron RS-35M repair
« Reply #10 on: September 17, 2022, 06:59:34 AM »

so with millions in use,,,whats a better supply at the same price...??
Logged

AA4PB

  • Member
  • Posts: 15504
Re: Astron RS-35M repair
« Reply #11 on: September 17, 2022, 09:02:23 AM »

So which would you rather have, an Astron with available schematics, parts, and plenty of information or a switching supply that the mfg. won't even give you a schematic for? I got rid of my Icom switching supply when I found that it wouldn't run on my battery/inverter system. The Astron supply works great and doesn't have any issues. And that scarry DC in the AC ground loop issue, run the numbers sometime. I did and in my system the DC current is less than 1mA when the Astron is putting out 35A.
Logged
Bob  AA4PB
Garrisonville, VA

AI5BC

  • Posts: 456
    • HomeURL
Re: Astron RS-35M repair
« Reply #12 on: September 17, 2022, 09:25:16 AM »

whats a better supply at the same price...??
Fair enough question. with several answers. My favorite two picks are commercial grade. First up is the Samlex SEC 1235M for $170 everywhere. This is a true 30-amp supply that will deliver 30-amps 24x7x365. If you do not want meters and save a few bucks, then the Samlex SEC 1235.Samels have several models. They all have UL  62368-1 listing, FCC Part 15B, EMI 55031 certification, DOE 10CRF, Caliphonies California Code of Regulations, Title 20, Division 2, Chapter 4, Article 4, Sections 1601 to 1608, and a 3-year warranty.

My second pick if you want battery operation is one of the many offerings from IOTA Engineering DLS series from 10 to 95 amps. For example, the 15-amp DLS15IQ4 model paired with a Trojan T-1275 battery makes an excellent combination for any Sammy Hammy. The Trojan battery you can take out to the field and operate all weekend without a recharge. Lime Samles comes with all the UL, FCC, DOE/Calaphony certifications, and 2-year warranty for $130. You can mount the PS under a desk out of site and out of the way.

Logged

K8AXW

  • Posts: 7391
    • HomeURL
Re: Astron RS-35M repair
« Reply #13 on: September 18, 2022, 09:31:40 PM »

5BC:  You make a very strong case and I don't doubt anything you say....except maybe for the "noise" part.  I've never had a noise problem with any Astron including the clones.
As for the rest, I guess this is what is meant by the phrases, "it pays to be ignorant," and "ignorance is bliss."

If and when my Astrons break I fix 'em.  When and if my Samlex 1235 goes belly up it'll be thrown away.

Thank you for the invaluable information OM.



Logged
A Pessimist is Never Disappointed!

AA7IS

  • Posts: 62
    • HomeURL
Re: Astron RS-35M repair
« Reply #14 on: September 19, 2022, 10:17:30 AM »

I don't work for Astron but hate to see them get a bum rap because of a few failures.

If you are willing to listen, let me point out details you have never considered or wanted to know. Let's start with why no ASTRON is UL listed. It has exposed live electrical components and unprotected metering wiring. All those exposed pass transistors have unregulated DC voltage, 19 to 23 volts depending on load. The collectors of the transistors are exposed and connected directly to the output rectifier/capacitor bank. Something comes in contact with the aluminum heatsink and collector, and BANG, dead short to chassis ground.

The panel meters have unfused and output voltage on them. Astron connected the panel wiring directly to the output without any current limiting device like a 2-cent 100-ohm resistor. They place calibration pots on the back of the meters rather than on the circuit board to protect the meter wiring—no UL cigar for Astron.

Grounding is a severe issue posing both safety and operational problems. Aside from the QC issue I pointed out earlier with the AC equipment ground chassis termination, the bonding of the chassis to the negative DC output is dangerous, destructive, creates RFI, and corrupts all and any grounding and bonding efforts. No other DC power supply manufacturer would do that because they know it does not comply with electrical codes and practices. All electrical codes and practices require all electrical systems to be Isolated from each other in the event of an electrical fault. If there is a fault (short circuit or high voltage), the fault current flows in all interconnected systems and can cause extensive damage to both systems. It is not rocket science, simple parallel circuit laws.

Hams compound the problem when they fail to comply with electrical codes. Hams intentionally place themselves in a ground loop, and your Astron makes it a DC ground loop. You run the coax directly outside to an earth ground (ground rod) rather than the AC Service ground before entering the house. Here is what hams do;  ground rod outside the shack to bond coax shield > coax shield inside to radio chassis > radio chassis to DC negative > DC negative to Astron chassis > Aston chassis to AC green wire in the power plug, power plug ground to AC earth ground aka service ground. You just placed yourself between two earth grounds with a serial daisy chain. You do not have a ground; your radio equipment is used as wire to connect two ground rods.

As a result, you get lots of problems you cannot fix or use tons of toroids trying to fix. With the DC negative bonded to the chassis, you lost your electrical and galvanic isolation between AC and Dc power systems, completely defeating the purpose of the transformer. All the noise currents in your home AC equipment ground system flow in your DC ground system and vice versa. You get a second DC negative wire. The one you know about, and the other is out on your coax shield to the ground and loops back around to the AC Service ground. You will have DC flowing on your earth ground system. What happens when you force DC current on ground conductors in moist dirt? Can you say galvanic corrosion destroying your ground system? Comply with codes and get rid of the jumper inside the Astron, and all that goes away.

Now let's talk about circuit design or the lack of it. I already addressed the voltage pot; however, another little tweak goes with it. If you cut the circuit trace to pin 4, the Inverting input and insert a 1K-ohm 1/8-watt resistor enables the Astron to be used as a Float charger, meaning you can connect a battery directly to the output with a fuse. No expensive high current blocking diode is needed.

The reference voltage amplifier has absolutely no filtering or soft start-start circuitry. A simple RC filter of a 3.9K ohm resistor and 10 uf cap placed between pins 5 and 6 give you an excellent soft start circuit and filtered voltage reference.
The LM723 power input is not regulated, resulting in poor PSRR. The LM723 has two power inputs: the regulator and the driver circuit. The driver circuit does not require a regulated voltage, but the regulator does if you want a power supply with decent PSRR. A proper design used a 78xx to supply power to pin 12 V+.

Current limiting and sensing are poor at best. Astron senses current via one of the pass transistor ballast resistors rather than a sense resistor on the output measuring the actual current. Inaccurate at best. Due to cutting corners and not using a series current sense resistor, Astron uses a resistor/diode network to simulate a current sense resistor to limit and control current, resulting in sloppy, inaccurate control. Compounding the issue is the 723 itself. If one looks at the Junction Temp vs. Current Limit, one instantly knows the 723 current limiter accuracy is poor.

RFI filtering, what RFI filtering? Astros lose regulation when used in solid VHF and UF environments like a collocated repeater site. Astron tried to address the problem 20 years ago by hanging an electrolytic cap across the output terminals. Any circuit designer or engineer with common sense knows that is useless because the ESR and inductance of the wire lead on the capacitor completely defeat the purpose.

The ICAS ratings are fiction. If you run the thermal analysis on the heatsinks clearly shows the rating is pure fiction. They get away with it because they know most hams do not push the limits. However, battery users, like repeater users, understand better. The first time your site goes on battery or places a discharged battery on an Astron without seriously reducing the current limit o 50% of the rating, it will require replacing the past elements. About the only thing most can do is run the output voltage as high as possible to minimize the thermal stress around 14.2 to 14.4 volts. Alternatively, if you know what you are doing is, add a 6/12 Vac buck transformer on the input to knock the transformer output voltage a couple of volts.

The bottom line is that the LM723 is an outdated late 1960s technology with high-side voltage control. It is inefficient, and when compounded by Astrons' poor design and QC, it is a poor choice when there are modern designs for less money carrying a warranty, UL, and FCC compliance.

For more reading, there is a website with a dedicated section on all the problems, fixes, tips, and tricks, along with every Astron schematic at Repeater Builder.
I've had plenty of Astrons over the last 20 years or so.
I use them because they're dirt cheap when compared to a 'REAL' commercial duty supply.
That is the only reason I've used them.
Most have required rework of poor soldering, poor application of conductive paste.
The overall quality of construction is something one expects from the cheapest power supply's on this planet.
I have a hard time imagining that Astron has "fan boys" but it seems within the realm of possibility.
Logged
Pages: [1]   Go Up