Manager - AB7RG
Manager Notes

The $10 Icom Computer Interface

Created by on 2000-12-19

The $10 Icom CI-V Computer Interface

Utilizing No External Power and all parts from

Click on the image to see a larger version


Note: On the miniature electrolytics, the negative lead is the

shorter lead and is marked with "0" in a white band



DB9 Serial Port Connections

The DB9 serial port pin 2 is the RxD and is applied to the LTC1383 pin 14

The DB9 serial port pin 3 is the TxD and is applied to the LTC1383 pin 13

The DB9 serial port pin 5 is a common ground in the circuit

The DB9 serial port's DTR (pin 4) and RTS (pin 7) supply voltage to the input lead 1 of the 78L05 5 volt regulator. You should be able to measure +5 volts output between lead 3 (output) and lead 2 (ground) of the regulator. Voltage is applied to pin 16 of the LTC1383.





Parts List



RS part #


Total cost


Linear LTC1383 5v RS232 to TTL converter IC





Ferrite Bead Pack of 25





22 uf 16 volt electrolytic caps





0.47 uf electrolytic capacitors





1N5818 diodes





1uH RF chokes





100pf capacitor





78L05 5 volt regulator





DB9 Female Connector

See text




1/8" miniature phone plug






Optional parts all from Radio Shack retail store: 16 pin DIP socket, aluminum case, metal standoffs, IC PC Board material (276-159B), grommets - all about $6 additonal.

Theory of Operation

The ICOM CT-17 RS232 computer level converter is both costly and requires external power. By using a cheap, readily available ultra low power TTL to RS232 converter IC, the Linear LTC1383, one can make a self-contained unit, which derives its power from the serial port itself. Although the IBM PC serial port was designed without a DC power pin, early mouse designers figured out a way to squeeze some low-current DC power from RTS or DTR lines (or from BOTH in the case of the design shown above). The LTC1383 draws only about 220uA at 5 volts - no more than most mouse circuitry. The above circuit should work with just about any serial port, including some otherwise finicky PC laptop ports. The LTC 1383 IC does all the work and requires only 4 capacitors to charge the internal pump circuits. Although one can use tiny 0.1uf non-polarized caps for the 4 charge caps, I chose some 0.47uf miniature polarized radial electrolytics to be extra sure of speedy response times in the heat of DX J . All parts are available at the time of this writing online from, but might be hard to find at the retail stores. I have no affiliation to Radio Shack, but enjoy their excellent website and variety of fairly cheap parts.

I chose to use some pre-punched general purpose DIP circuit board, a 16-pin DIP socket, 4 standoffs, 2 grommets and a tiny aluminum case to wire the circuit - which would add about $6 to the cost. Anal-retentive hams can etch and drill their own circuit boards! I'm sure some industrious hams will figure out a way to package the whole circuit inside a DB9 shell!

Instead of purchasing a DB9 female connector, I pilfered the connector and cable from an old MicrosoftTM serial mouse. I not only got a "free" DB9 connector, but I found that the 4 conductor shielded cable used by the mouse already had my 5 connections made to the right pins! I even used a couple of feet of the same cable (after cutting off the 3 unused leads) as a single conductor shielded cable to run from the aluminum box to the 1/8 inch miniature phone plug for my IC-746.

IMPORTANT: If you want to use a DB25 serial connector instead, remember that the ground pin is 7, DTR is pin 20, RTS is pin 4, TxD is pin 2 and RxD is pin 3! This would certainly give you more room in the casing to house the circuitry if you decide to go that route. I understand that the same type of circuit can be used for a Kenwood interface if you use separate leads to IC pins 11 and 12 (not tested).

Questions, comments, photos of your creations, etc. are welcome to Although I built a working unit from these instructions, I make no claims or warranties about the suitability or correctness of these circuits.

I am not an engineer and I don't play one on TV.

Copyright, 2000, Howard Groveman KD6UU

IZ4FHT 2003-11-25
RE: The $10 Icom Computer Interface
I have moved my homepage and the interface diagram is now at:

Andrea (IZ4FHT, formerly IW4EGQ)
IW4EGQ 2003-01-14
The $10 Icom Computer Interface
The mouse cable I recycled only had RTS connected so I decided I'd give a shot to a very low power solution: a MAX3232CPE (as suggested prviously) together with an LP2950.

Here is the revised schematics (with many thanks to the original poster and to all previous contributors):

Some of the following comments stem from a discussion in it.hobby.elettronica:

-do not use a 78L05: no sense in using a low-power IC as the MAX3232 or the LTC1383 if the voltage regulator wastes 3-5mA just to operate (not to mention that it requires at least 7V to work at all).

-limiting resistor R1 in is probably excessive: 1Kohm should be enough (as suggested by Mario). But *do* add a resistor and the diode! Mario's mod is important.

-probably, ferrite beads alone are enough, no need to add 1uH inductances, but better be safe than sorry, right?

- ferrite beads (or a single clip-on choke) should go on all serial lines, not just on power sources. Reasoning is: either trust all lines to deliver no noise, or none of them.

The two key components are slightly more expensive than plain vanilla alternatives, but this should really be the most energy-saving solution ;-)

Have fun,
Andrea IW4EGQ.
KC2FCR 2002-07-08
RE: The $10 Icom Computer Interface

Short answer is that since the original circuit (NICE design, btw) stole power from the computer port to power the interface, you'll need a source of power on that side if you opto-isolate. Not a mod, it would be a redesign. My suggestion would be to let everything share a common ground (Yeah, I know, people screaming in the background), or put a couple of .05 amp pico fuses in the ci/v lines.
WB7G 2002-05-12
RE: The $10 Icom Computer Interface
To do PSK31, I built a computer to rig interface with optical isolation because my computer is on a different wall outlet, but now, this computer to rig (for computer control) has no opto isolation!
This is scary if the grounds are not the same. How do I add isolation to the circuit above.

Use a 4N33 on the logic side? Confused because the
input and outputs of the LTC1383 are tied together.
(after the diode and resistor if you add the Mario
mod which I am.)

Can someone send me an ckt. example ?

Thanks ahead.

KD6UU 2001-01-18
RE: The $10 Icom Computer Interface
You didn't reduce the internet pileup - the article only stays on the eHam home page for 1-2 weeks. It then fades off into ham oblivion and you would have to hunt for it to find it. Only the author (and management I suppose) can see the number of page views and it has been interesting to follow. They pay an author $1 for 1000 page views, so I am waiting a while before I put in the tennis court and indoor pool :)
S5M 2001-01-18
RE: The $10 Icom Computer Interface
I didn't realize I reduced Internet pile-up here :-)
LT1383 under "Any receiver output short-circuit current" has typical sinking figure of -40 mA.
This would be too much for other CI-V drivers.
CI-V operates at +5V level and I would not recommend 3V ICs although they can usually withstand higher voltages. MAX232CP consumes 5mA at 5V. Compare this with any HF XCVR which needs 2A at 14V for RX only.
The smartest ICOM mod I've seen lately is bridging separate CI-V (or remote) jack to ACC1 socket...
I stil wonder why I haven't done it millenium ago :-)

73 de Mario, S56A, N1YU
KD6UU 2001-01-18
RE: The $10 Icom Computer Interface
Mario, thanks for testing the circuit. I wish you had done this in your very first note. When you said "LTC1383 won't work" we were getting 1000 page views per day. Now, we are down to 15 page views per day. But at least you are being a scientist. I have heard from many sources that you know your stuff, so let's keep the dialogue going. As long as you are considerate, I think that everyone will enjoy learning something new from you.

Now, where do you get the 40 ma short circuit current spec? Here is the data sheet on the LTC1383:
Also, can you do me a favor and look at the MAX3232CPE. This chip looks intriguing and it only requires 3V at very low current. It is a pin-for-pin replacement for the LTC1383 and could probably use a 3V regulator such as the 78L03:
Thanks, Howard KD6UU
S5M 2001-01-18
RE: The $10 Icom Computer Interface
I just made MAX232CP interface in order to verify KD6UU design. Tantal 10uF caps were used with +5V stabilized power supply. Ground short on CI-V side produced ONLY 6 mA which is well within ICOM specs.
This current was drawn directly from +5V supply without affecting +/-10V RS-232 interface voltages.
Maxim claims 10mA as typical TTL/CMOS output short circuit current while LT-1383 has much higher 40mA figure. I'd appreciate, if someone can measure CI-V short circuit current of KD6UU design. It might even work reliably as proposed!

73 de Mario, S56A, N1YU.
S5M 2001-01-14
RE: The $10 Icom Computer Interface
ICOM sells significantly more VHF rigs and therefore suggested 1uH/100pF blocking. Their HF xcvrs are blocked with 100uH/10nF.

73 de Mario, S56A, N1YU, nothing to do with NY :-)
KD6UU 2001-01-12
RE: The $10 Icom Computer Interface
Mario - so far, you have only been ridiculing the left side of the schematic. Start working on the right side, OK? The 100pf caps and 1uh chokes were suggested by Icom, so take it up with them. Maybe you have been affected by the CO?
S5M 2001-01-12
RE: The $10 Icom Computer Interface
Recent experience of Kuby, N6JSX/8 with notebook and switched RTS and DTR low voltage signals should be of great concern because 2 x 22 uF capacitors need a lot of time to charge. One might consider 13,8V from radio as an external supply available when needed.
BTW 100pF blocking capacitor is too small for HF...

73 de Mario, S56A, N1YU

P.S. Howard, I learnt more about your job while saving XYL from CO poisoning in the kitchen!
KD6UU 2001-01-10
RE: The $10 Icom Computer Interface
Thanks for the follow-up Kuby. I hadn't thought of the new 3 volt notebook generation I guess. You may be able to just change the regulator to a 78L03 3-volt deal, since the voltage range is pretty broad on the LTC1383. There are also plenty of circuits out there that don't even use a regulator. The new MAX3232 is a 3 volt IC but tolerates over 5 as well -so I hope someone tries it and gets back to the thread. It might be even more foolproof than the LTC with a 78L03 regulator for those with notebooks. 73, Howard
N6JSX 2001-01-10
The $10 Icom Computer Interface
FYI, a few facts that you may need to know....
Digi-Key usually has a $25 minimum on orders.....

I built the above circuit and came up with a problem, however, I suspect it is due to being used on my NEW Sony Viao Notbook PC.

The voltage at DB9 pin 4 and 7 is only 5.3 VDC coming from the PC. So I bypassed the voltage regulator (for a lil while). I have a number of MAX232e chips so I used a Maxim chip. My Viao did not supply enough current on pins 4 & 7 to get the MAX chip operational.
I ended up installing a locking toggle switch with a 9V battery and then used the regulator part of the circuit - it NOW works just fine. I was in a rush so I did not install a power LED - which forces me to watch which way the switch is - or I'll have a dead battery in a few weeks.

I was using this circuit in conjuction with the Kwood V7a program. The program or Viao - ONLY supplies power to the DB9 when a "read" or "write" memory is commanded, so don't get worried if no voltage is present in the dormant "not-being-used" serial port mode.

Kuby, N6JSX /8
KD6UU 2001-01-10
RE: The $10 Icom Computer Interface
Sorry -just got an e-mail from a builder that the above sample Digikey order got wiped out. If you are ordering from, here are some Digikey part numbers:

LTC1383CN (LTC1383 IC $5.63 ea)
M2306-ND (Ferrite Beads $0.25 ea)
1N5818GICT-ND (The 1N5818 diodes $0.45 ea)
78F1R0K (1.0UH choke - $0.55 ea)
296-1365-ND (The 78L05 Regulator - $0.56 ea)
P4925-ND (100PF 50V CERAMIC) $0.38 ea)
P808-ND 22UF (16V MINI ALUM ELECTROLYTIC - $0.28 ea)
P5173-ND (.47UF 50V ALUM LYTIC RADIAL - $0.21 ea)

Good luck, Howard
KD6UU 2001-01-09
RE: The $10 Icom Computer Interface
One more thought on the above Digikey info. If you put in the above web ID and access code, print out the sample order I put in and then re-log in to Digikey and enter your order from scratch (be sure there is a new order code). I just had this scary feeling that someone would enter their order and credit card info my "sample order" and then that info would become public domain when the next person goes to look up the above web ID and access code. Can't be too careful. Howard
KD6UU 2001-01-09
RE: The $10 Icom Computer Interface
Nice work, Karl! You can actually get ALL $11 worth of parts on the site. If you need other things and can spend $25, they waive the $5 "handling fee". For others looking for an alternative to, try Karl's suggestion: and click on online ordering. I put in a boilerplate order which you can call up to place your order. Enter Web ID: 614757 and Access ID: 41071. This should bring up a nameless $11 order which you can then customize. I think the parts will be perfect, but the descriptions are scant, so don't call me if you get a 2 inch diameter electrolytic in the order :)

The only other things you might want are available at the Radio Shack store - aluminum case, grommets, a DB9 female and shell if you need, 1/8" (3.5mm) miniature mono male phone plug, metal standoffs, and IC PC Board material (276-159B) - about $6 or $7 more.
KC0JHB 2001-01-09
RE: The $10 Icom Computer Interface
Howard, I found the LTC1383CN at Digikey Corp. 1-800-344-4539 or "" They have them in stock for $5.63 plus $5.00 for handling. Thanks again!

KC0JHB 2001-01-08
RE: The $10 Icom Computer Interface
Howard, Thanks SO much! I will give it a try and I will also let you know how I come out. Thanks so much, your information may help others as well. It seems this post has put a run on the Radio Shack stock. :)

KD6UU 2001-01-08
RE: The $10 Icom Computer Interface
Karl, I called Radio Shack and they said they would tell their buyer to get on the stick.

But all is not lost. Here are some options:

If you don't mind working with the surface mount pins, Radio Shack is not out of stock of #900-6408

You can try Linear directly: ask for the LTC1383CN - Linear Technology Corporation, 720 Sycamore Drive, Milpitas, CA, 95035 Tel: 408-432-1900 Fax: 408-434-6331 e-mail: . They are supposedly going to begin selling online next month. You can try for a sample :)

Maxim has come out with a competitive chip that they cross-reference to the LTC1383. I haven't tried it, but I would bet the ranch that it works. The MAX3232CPE should be a direct 16 pin DIP replacement and it is only $2.86. The Maxim website didn't like IE 5.5 - I had better luck using Netscape to get to their order page. You have to register with a valid name, company name and address in order to place an order:

Let me know how it goes. Howard
KC0JHB 2001-01-08
The $10 Icom Computer Interface
Radio Shack is out of the P/N 900-6407, Linear LTC1383. Does anyone know where to find this other than Radio Shack?

WB5VCO 2001-01-04
RE: The $10 Icom Computer Interface
Maybe I have missed it but where can I find software to control a 728?

KD5CTJ 2001-01-01
The $10 Icom Computer Interface
Just put the circ together and it works as advertized, minus a few burnt fingertips! hihi. Thanks to Howard, KD6UU for posting this.
73-Erik, KD5CTJ
KB3CO 2000-12-30
RE: The $10 Icom Computer Interface
Regarding Pat's (N0HR) post:
The RIGblaster does NOT provide an interface for the receive path - only the transmit path.

KD6UU 2000-12-28
RE: The $10 Icom Computer Interface
Don't want to beat a dead mouse, but I have compiled a lot of the information, Q&A, an updated parts list, suggestions, corrections and yes, even my friend Mario's circuit mod on the following site which you might want to review if you are about to build this interface or if you are otherwise interested in the topic:
Thanks to everyone for their feedback and assistance!! It has been more fun for me than breaking through the biggest pileup on 20. Howard KD6UU
N2NL 2000-12-27
The $10 Icom Computer Interface
FYI, the parts I ordered arrived and I built the interface without any troubles - it works great with my IC-706 - no problems at all. I did not add the modification Mario suggested because I didn't have the parts on hand - but it works fine without it.
Although I've never met Mario in person, I've followed his posts over the past years and know him to be extremely knowledgable. His recommendations deserve consideration. I feel that in everyday, normal use (1 rig, 1 computer) the original design would work fine as it does for me. Mario's recommendation probably would fix possible problems that might appear when daisy chaining ICOM rigs or other complicated setups.
We must realize as hams who use English as their first language, there are many others in the world who aren't as proficient and who's point may come across differently as they intend. I'm certain that Mario wasn't intending to insult anyone.
Finally - thank you Howard for posting this article - it was a fun project to build and it saved me over $100 over the cost of buying Icom's CI-V.

73, and Happy Holidays,
Dave N2NL
(formerly KH2/N2NL)

PS - the ferrite beads - part number 900-5005
at - are EACH - not for 25 - so make sure to order 2 of them
S5M 2000-12-26
RE: The $10 Icom Computer Interface
DR OM Howard, let me give you an example you surely understand better then I do: My 8 year old dog was recently diagnosed by diabetes. Was he healthy dog?
Just like your circuit "works"...

I did read Kuby article on grounding and it help me with Polyphaser reference. No further comment as I feel the issue of RF ground is not well covered.

My beloved comment about humans should not be taken within this technical content.

If I write a nice article that Earth is flat, what kind of comments should I expect?

Respectfully Yours

Marijan Miletic, S56A, N1YU, mother tongue Croatian.
KD6UU 2000-12-26
RE: The $10 Icom Computer Interface
One point to those ordering the parts from Someone mentioned that when they ordered the ferrite beads, part #900-5005, they only got a single (1) bead instead of a pack of 25 like I did. For the extra $0.25, you might want to order two to be sure. Howard
KD6UU 2000-12-26
RE: The $10 Icom Computer Interface
OK, now I think I fully understand what is going on with Mario. I believe I was gracious to him and tried my best to fully understand his point since he told me he was an RS232 engineer. I took the time to review the references he e-mailed me (Again, I not an engineer, but a family physician by training) and only wanted to pass on a simple circuit that had worked for me. I still don't really know if his 2 additional parts will make my circuit "electricly" correct or whatever. Any circuit can be improved upon - so what is correct anyway? I will simply state the truth as I know it. My original circuit worked for me and the others that have written to me. It worked with every piece of software I tried - with FULL bidirectionality and with no software errors that I can detect. If for some reason it doesn't work for others, try Mario's two extra parts :). And as for you Mario, I don't think I have met a more passive agressive ham. If you have fountains of knowledge, write your own articles and comment constructively on the articles of others. I really didn't appreciate the quote about stupidity or your snide statement about sacrificing an old worn-out serial mouse. Kuby really had your number. Like me, he also took the time to submit an article to eHam (see the excellent article "From the Ground Up" he wrote) He most certainly deserves an apology. And if you don't have something constructive to say on this thread, please give us all a break and stop cluttering it up. Howard
S5M 2000-12-26
RE: The $10 Icom Computer Interface
Kuby, N6SJX/9 asks: What is the MAX. current that the CTS/RTS can provide?

RS-232 specs allow +/-15 V max and 300 Ohms TX resistance min. I leave the math to you or FLA :-)

73 de Mario, S56A, N1YU
S5M 2000-12-26
RE: The $10 Icom Computer Interface
The main reason that certain software packages may NOT work with KD6UU interface is that they do NOT force either DTR or RTS RS-232 COMM port signals so there is no power supply. It can be easily measured but who cares these days when you cut a mouse just to get a cable :-) Or consume 25% of energy for 4% of people...

73 de Mario, S56A, N1YU
S5M 2000-12-26
RE: The $10 Icom Computer Interface
I did write several times what is electrically wrong with KD6UU circuit and offered proper solution.

Please check in ARRL Handbook about Open Colector drivers. These are required for CI-V shared line as they are made to sustain multiple shorted outputs. LTC-1383 has CMOS output stage which will consume 40 mA when shorted. MAX-232 is the same but who cares about the IC internals these days as long as it "works". M$ did teach us to live with a software problems...

MFJ owner said it nicely: Hamradio is the knowledge of usin appliances. My good friend DL7AV calls it
"steckdosse" operators.

73 de Mario, S56A, N1YU MSc EE
N6JSX 2000-12-25
RE: The $10 Icom Computer Interface
Enough is enough, Mario. You profess that the circuit is "electricly wrong", even though at least three sources say it works - and works well.

I find it interesting how some people can take the "time" to critique and criticize an artilce but do not offer their vast wisdom for ALL of eHAMdom to scrutinize. If this circuit is so wrong - eductate "us" (the ignorant) and submit an "electricly correct" circuit schematics so it too can be scrutinized by all of eHAMdom! Otherwise, your beginning to sound like HOT air.

And don't take it personal when your submission, that took you hours to compile, is tore up by someone hiding behind their own super knowledge or "Anonymous" posting. It's all in fun!?!?
S5M 2000-12-25
RE: The $10 Icom Computer Interface
Ed, AF2P measured LT-1383 current consumption with pin 14 RXD short. This is RS-232 output which is always protected by 300 Ohms internal resistance as external cable shorts frequently occur. You can try to short CI-V connection in order to verify manufacturers data but I am NOT advising you to do that. Original circuit is electrically wrong but the errors would show during PC data reception from ICOM CI-V line.
I own one of the first IC-735 which introduced CI-V.

Didn't you hear about initial Pentium bug? Only 5 bytes were missing in milions...

73 de Mario, S56A, N1YU

P.S. Stupidity is a cosmic force :-)

KD6UU 2000-12-25
RE: The $10 Icom Computer Interface
Happy holidays everyone,
I've heard from several successful builders now - so this circuit has been a lot of fun to share. Eham has an incredible following - 3000 viewers of this article so far! What is really nice is the interaction following the publication of an article. I have been in touch with Mario (great guy who has a lot more experience with RS232 interfaces than I have) by private e-mail and he suggests the following circuit mod which I have not tried or tested: see my rendition of his mod at
This is about a 60 cent parts addition - seems easy enough to include it. He chose a 10K resistor and another 1N5818 diode after further study since his original comments above. He feels that this will prevent the possibilty of software errors if multiple Icoms are connected along the CI-V interface chain and too much current is drawn through the interface. His concern is apparently NOT an issue of possible "damage" to Icom, computer, or interface - just possible software errors due to this theoretical possibilty. My point was that the interface above runs VERY low current -it can't pull 40 ma if shorted as Ed found out :) ... And that it was Icom who decided to tie the input and output TTL line together in the first place! Besides, I haven't seen any other CI-V interface that have used these extra components. But Mario certainly knows his stuff and recommends it. So if anyone else is building this interface, give it a try and let us know how it goes. Best 73, Howard
AF2P 2000-12-24
RE: The $10 Icom Computer Interface
Where is the 40 ma drawn from? Are you indicating that 40 ma is drawn out of the
remote port on the ICOM - or are you indicating that the interface draws up to
40 ma from the serial port, under certain conditions?

I just built Howard's circuit - works great! I measured the current draw, using
a power supply rather than DTR and RTS and measured 2.8 to 2.9 ma.
My final version uses the serial port as the power source. Re-wiring it to
include the resistor and diode will be a bit difficult. If you really think it's
necessary, I'll do it. The LTC1383 seems very hardy. When I put the
thing together, I had a dead short from pin 14 of the IC to ground, whenever
the serial cable was installed in the computer. The (*^&$ mouse DB9
connector had a short from pin 2 to the metal case of the DB9. I
messed around for several hours trying to find the problem, until I thought
of the metal case of the DB9. Any way, the chip survived, and the measured
current draw with pin 14 of the IC shorted to ground was 21 - 23 ma with
the powersupply connected - it dropped the voltage down to 2.75 on pin 16 of the IC
when power was taken from the serial port. With the short cleared, the
current dropped to between 2.8 and 2.9 ma and the voltage came up to

Hope you have a great holiday!

S5M 2000-12-23
The $10 Icom Computer Interface
You should add 1K Ohm resistor bridged by 1N4148 diode with cathode on pin 12 to CI-V connector in order to limit the current consumption of other ICOM radios to 5 mA when they send packets using common CI-V line.

You may go up to 40 mA peaks with original circuit during CI-V low logical levels. Lower curent limitation probably occur due to RTS and DTS feeding capabilities but 22 uF doesn't help.

Errors caused would be falsely blamed to software!

This circuit was published on SCC reflector and attracted my response. I've been interfacing radios to computers via RS-232 since 1983. MC-1489 is my best choice including KEY and PTT lines.

73 de Mario, S56A, N1YU.
KD6UU 2000-12-23
RE: The $10 Icom Computer Interface
Mario - one more thing regarding your comment about the LTC1383 chip. Perhaps you haven't set your Icom to "CI-V Transceive" mode. This would only allow one way commands from the PC to the radio. Using AA6YQ's excellent freeware called "CI-V Commander", I verified that the LTC1383 is sending commands in both directions without skipping a beat. In fact, anyone building the interface should try this software to verify that they built the interface properly and that their Icom's CI-V address, baudrate, etc. are set correctly: . It is also a great tool to learn about CI-V commands. In the case of the IC-746, the radio defaults to CI-V transceive, address of 56, and baud rate is "Auto". All the software I tried can be set to any baud rate up to 19,200 and the Icom recognizes the rate automatically and the data works in BOTH directions. That said, some controls/features MUST be used on the radio OR the software - for example, turning up the AF on the computer doesn't rotate the AF knob on the radio :) Howard
KD6UU 2000-12-22
RE: The $10 Icom Computer Interface
Mario, thanks for your comment. But please explain in more detail. As in my circuit above, I have pins 11 and 12 "shorted" together and connected to the CI-V on my Icom 746. I have noticed no lack of control from PC to Icom or from Icom to PC. And I have verified this with no less than 4 pieces of CI-V software. I can tune, change modes, and change bands freely from either and they track perfectly. Where should I be looking for the non-reliability? Are you saying that a direct substitution with a MAX232, its pin-compatible older cousin, would work better and more reliably somehow? If there is good reason to use that older IC, I'm sure those that are building the interface would like to know now :). Thanks. Howard
S5M 2000-12-22
The $10 Icom Computer Interface
LTC1383 used does not allow short on CI-V side and would not work reliably on receiving data from ICOM.
Sending data from PC to radio is OK.

Happy holidays de Mario, S56A, N1YU MSc EE
N.A. 2000-12-21
The $10 Icom Computer Interface
Here is a list of some Windows Rig Control software to use after building the interface. Try the 746 software even if you don't have a 746 - it is stunning!

1) IC-746 software can be found at:

2) Software for the IC-706, 706MkII, and 756 can be found at:

3) TRX-Manager offers a free (time/use limited) demo and works with most Icom models:

4)YP Log Program (Logging with some feature/frequency control):
N6JSX 2000-12-20
The $10 Icom Computer Interface
FYI, an added note to this design.

If you intend to do PC sound card PSK or SSTV you will wnat to substitute the DTR connection with CTS. As most sound card programs use the RTS or DTR line for hard PTT keying control signal, ie, W95SSTV, WinPSK, and more.

Making this minor change will allow you to use the DTR with these type of programs for program controlled hard PTT keying! In my curcuit I use a opto-isolator to the radio for complete signal isolation.

73, Kuby, N6JSX /9
N0HR 2000-12-20
RE: The $10 Icom Computer Interface
For those who would like to avoid a soldering iron as much as possible, yet want a computer interface with their Icom rig without spending (relatively) TOO much:

Rigblaster (from which handles all of the TX/RX<->soundcard interfacing EXCEPT for CW keying. ~$90

The LCU-3 is an enclosed converter that performs just like Icom's for ~$49.

A keying interface is $25.

You'd then have everthing needed.

While I do not represent any of these companies, I can vouch for the LCU-3 and Rigblaster. However, for keying, I built the simple 4n33 optocoupler circuit:

I'll recommend against using any keying circuits which rely on the parallel port (as programs like WriteLog don't support LPT keying under Windows 2000).

Along these same lines, another bit of advice. Use software that allows CW keying and rig control on the same serial port (as WriteLog does). Then use a "Y cable" to run CW to the keying circuit and rig control to the rig control circuit. Also, when you do run out of free serial ports (for mice, rigs, TNCs, modems, etc.) make sure you expand with a card that allows IRQ sharing, otherwise you'll run out of IRQs as well. I found a ~$48 PCI slot card that adds two ports but is expandable to four (for a total of 6 in my machine). Works great:

Anyway - I'm not selling anything here. Just giving a quick and dirty way to get up and runnin'.

GL es 73,
KC7YCL 2000-12-20
The $10 Icom Computer Interface
I'm lazy, does anybody make these in pre-assembled or kit form? I'm looking into getting a 706 ($904 at!!) but the accessory prices are incredibly high.
W3EEE 2000-12-19
RE: The $10 Icom Computer Interface

The '701 predated Icom's CI-iv and CI-v serial control systems, so it's unlikely any code for those will help you at all. To the best of my memory, the '701 doesn't even have a microprocessor in it, but the big fat Molex connector on the back actually allows you to jam straight into the synthesizer's counters to set the frequency. Toyed with the idea of knocking up a bit of hardware to interface from a printer port once in the dark ages, but . . .
Connector pinout and the somewhat arcane methodology are in the manual; I've probably still got a copy lurking around somewhere if yours has gone 'walkies'.

Cheers, Steve
KD6UU 2000-12-19
RE: The $10 Icom Computer Interface
In answer to a couple of questions posted here and sent to me by e-mail:

The serial port should be able to provide a good 5-10 mA. The 78L05 regulator needs a couple of mA to work. A MAX 232 has been used in the past very sucessfully (or a MAX 233 which I think doesn't even require the charging caps but has a different pinout I believe, so be careful). The LTC1383 is ultra low power and CHEAP and available from Radio Shack - so no downside, and it works great.

As for programming the CI-V interface, Ekki DF4OR, is one of the worlds experts. He has a great site at Lots of info and other CI-V interface circuits besides mine.

As for which rigs this will work with, I believe the answer is found in your manual. If the CT-17 is listed as an option, this circuit should work. Rigs like the IC-761, 765, 756, 775, 781, 746, 271, 471, 275, 475, 725, 726, 728, 735, 736, 737, 970, R71, R72, R700, R7000, R7100, R9000. Someone said that the circuit will work with Ten Tec DELTA II, ARGONAUT II, OMNI VI, & PARAGON. Some older Icoms like the 751a required a UX14? option since they were CI-IV and not CI-V. I don't know about the 701 or the use of this interface with handhelds. Maybe someone can describe the interfaces used for cloning/programming handhelds - do these use computer to audio encoding or are they TTL too??
KB3EHW 2000-12-19
RE: The $10 Icom Computer Interface
There is a control program listed in the linux ham software database at Look under rigcontrol. I havn't tried this software, but it looks like the correct thing.
If only I had enough serial ports... :)
N1YRK 2000-12-19
The $10 Icom Computer Interface
Great article!
Can anyone help with the software part? I have an old IC-701. I am a Linux user and would prefer to not have to install windows just to use my radio. I am a halfway decent programmer so if someone can provide the details on the control protocol I can write the program, if none exists. And I'll be happy to add support for other radios too, if I can get some volunteers for testing.

N6JSX 2000-12-19
The $10 Icom Computer Interface
OUTSTANDING article! Nice to see some good technical articles rather than opionion articles.

I have designed nearly the same circuit except that I used a MAX202/232. I've not built it yet due to one issue that I've been trying to get information on. What is the MAX. current that the CTS/RTS can provide?

I have built a very simple PNP, NPN 5V zener circuit that works just fine and recently found a posted article on the WEB by KC7ZRU that reversed engineered his TM-V7A interface circuit. I found my circuit to be almost the same design.

The simplicity of these designs is that NO external power source is required! I'm working on a slightly improved interface circuit that will incorporate your basic RS-232 to TTL design AND have isolation transformers and optoisolators between the PC sound card and DTR signal and my Yaesu HF rig so I can do PSK, SSTV, etc. I want to keep the PC electrical circuit as isolated as possible from the radio - ensure no ground looping/humm can occur.

TNX, Kuby, N6JSX /9