I have all the semiconductors now, except for the bridge rectifier, which should turn up any time now, and the 4700uf and 470uf electrolytic caps. Once I get all the parts, I will replace all of them in the PS, and see what happens. If that fixes it, great, the next thing will be a large scale replacement of all the electrolytic caps in the R71a. If I go that far with it, I probably will just go ahead and to the audio mod to it, and keep it, as it's in nice physical shape. I may add a fan to it to keep the PS cool too. The front panel will probably have to come off to really be able to get it cleaned. I've done that a couple of times before, so it's no huge deal, just a hassle.
Replace all the lytics in the PS section as well as the DC-DC converter as dkf435 mentioned.
The DC-DC converter is often overlooked, especially in the R7000. I don't recall if the R71's had the same DC-DC converter board. I've not had mine open for years now.
I'm also a fan of powering the older Icom's from an external power supply. That reduces the heat inside the chassis by a significant amount and can only help prolong the life of the other irreplaceable components like the CPU.
If you must use the internal supply, it is not a hard mod to fabricate up a small 12 volt fan of say the size that cooled the old 386/486 CPU's to help pull or push air through the chassis. That does introduce dust into the radio though. The dust and noise reason is why I use an old 40 amp external supply for most of my radios. I use the supply to maintain a large SLA battery so I can operate during power failures. I also used an aircraft landing light as a current limiting ballast resistor so I don't fry the battery should AC fail and run the SLA down when not home. The landing light limits the current inrush into the SLA when AC power returns. My external supply does not have current limiting ability so the landing light used as a current limiting device worked out perfectly. Maybe some day I'll install a load sensing/limiting circuit but I've been saying that for years now!
I also have an Icom R-9000 that works on external power but part of its metering circuit is dependent on a stable input voltage independent from the rest of the receiver. When I run the R9000 from an SLA that is discharging, the S-meter will start giving false readings. I think it was mainly the center tuning function that would no longer give accurate readings as the SLA started discharging. The R7000 may also have that problem but I rarely use its center tune meter setting. Not sure if there is anything in the R71 that expects a set input voltage. Worse case is usually the s-meter may read a bit low and maybe dimmer backlights but the radio will still receive just fine if your external battery is dropping below the power supplies setting. Most of the radio circuits are regulated down to 5, 8 or 9 volts with the DC-DC converter handling the odd voltages needed to run the display mostly. The audio output may also drop off some as that stage is usually powered from the primary regulators setting.
For the R9000, I opted to use its internal supply with a large power supply fan mounted on the rear mounted heat sink. The fan keeps most of the heat out of the radio as I also did some tricky ducting to pull a small amount of air through the chassis. I also have the 9000 on a large UPS which will run it for about 14 hours.
All of my older Icom's had dried out caps. Most were in the power supply stage or in other areas that ran warm.
Not a single one of them had a cap that looked bad to the eye at all but they were indeed bad.
Since recapping them all years ago and then later switching most to an external supply, all of them have worked flawlessly and some are on 24x7 for days or weeks.
I never swapped the back lighting with LED lamps so I usually run the lights in the dim position to give them some added life. I did swap the meter light in the R9000 with a small LED.
I've also replaced the CRT in the 9000 twice and have an LCD module for the next time the CRT starts to burn in images as it spends a lot of hours on the same frequency.
The R7000 and R71A have both had display issues. I don't recall the cause of the R71's display issue but the R7000 was caused by bad caps on the DC-DC converter board and later, the well known failure of the ribbon film type cable that connects the display board to the logic board.
Otherwise, I can't say I've really had any other component failures in any of my Icom's. A few mechanical issues with one being the power switch and audio jack failures.
The R7000, R71 and several others use the same power switch.
My switch failed on the AC internal supply power contact side of the DPST switch. It would drop the AC input voltage nearly in half when it would fail. Turning it off/on would sometimes restore full voltage but it would fail again sometimes within seconds. My power switch failed in both my R7000 and R71A. Identical failures of both.
So check the power switch voltage across its solder joints or at the molex connector located on the wiring harness between the switch and the power supply section if you still see the voltage drops after replacing the caps.
One side of the switch switches the unregulated DC voltage and the other set of contacts switches the AC mains side before the power transformer.
A quick fix is to short the DC line and remove it from the switch and then move the AC lines over to the now unused DC contacts as the AC side is usually the set of contacts that will fail.
Not an ideal solution especially if you intend on running from external power but it works until you can find a new switch. Icom still sold the exact switch but wanted something like $40 dollars for one a couple years ago.
I found an identical switch made by Alps, I think, for $3 dollars on eBay.
I bought several but have never had another fail. Maybe because mine remain powered on so the switches are rarely used?
So, replace the caps and also check the voltage across the main power switch when your voltage drops in half. Do that and I bet you will have a nice radio!
A bad power switch can also usually be detected by depressing the switch just a small amount and see if voltage returns. Don't press it enough so it pops out when released, just press it enough to move the contacts.
If the switch is bad, you will usually see the voltage return to normal with just a small press of the button or at least see significant voltage fluctuations. Small fluctuations are normal even for a new switch. Fluctuations large enough to see visible changes in the backlight level is often a good indication that the power switch is suspect and should be checked.
I also did a live coin cell replacement in my R71 before it died. I still ended up buying the Piexx ram board for the R71 as it extends the tuning range over the stock Icom board.
The factory lithium coin cell used on those ram boards can easily remain good for over 20 years. May not be true for a radio that has been sitting unpowered in a closet for years but for the radios that spend a lot of time powered up, 20+ years is not out of the question. That is thanks to the small amount of reverse leakage of the diode that prevents the coin cell from being charged. The cells are not rechargeable cells but the small amount of leakage through the diode does keep them charged without damage to the cell and is what is responsible for the 20+ year life many owners see. Heck, I guess it is now over 30 years life and many of them are still working fine if they spent a lot of that time powered on.
The factory ram boards are also user reprogrammable if one wants to build up the interface needed but I think it took an older slow computer running DOS and may have been through a true parallel port if my memory is correct. The code is available on the internet. It is much simpler to by a Piexx or similar board or do a live battery swap as long as you are careful and don't short something during the swap. All the 3rd party boards hold the OS code in non volatile ram but some do still have a coin cell for holding the memory channel contents or maybe RTC settings. Most also offered extended tuning ranges and some offer more memory channels. The extended tuning range is dependent on the radio model. For the R71, I can't use all of the new tuning range but can use a good part of it. I think the Piexx board extended the tuning range of the R71 from 10 kHz up to 33 MHz. Sensitivity dropped off pretty fast when I tuned much below 80 kHz I think but it worked fine all the way up to the new upper limit which I think was 33 MHz with the Piexx board. Of course about everything above 30 MHz was useless unless you have the FM option. I think the Piexx board also had a jumper that you soldered in or removed that retained the factory tuning range if desired. It also had solder pad jumpers that must be set to the radio model you installed the board in. The same ram board with OS code was used in several Icom models from the same era.
