ft847 on 4 metres

Questions and answers on 4 m transceivers, transverters, antennas, etc.

Postby G0KSC » Sun 29 Jun 2008, 11:09

Nige-G7CNF wrote:Andy.

I have written a long scrawl about this in a snail mail letter to some amateurs with no internet connection. Basically the permeability of the core is critical for this application. In my letter I expressed concern that there may be some less experienced amateurs with little or no previous engineering or homebrew experience who will have undertaken this mod blindly, using cores from 70/10Mhz/455kHz IF stages.

These cores will get exceptionally hot due to the high flux density present in the centre of L5006 and there will probably be a few 847's knocking about which will start to rattle soon!

This is why Keith used the Toko VHF/UHF core, which has a relatively low mu compared cores found in lower frequency IF stages, which is where I suspect many of the ferrites will have come from that have been installed in the rigs. If you have an LCR meter, bridge or SWR analyser, it is possible to check your available cores and roughly calcualte the mu, this would save much heart-ache. What may be harder to determine is the impedance @ 100Mhz, which is also a critical factor, as this quality will determine the level of heating.

The only satisfactory way to complete this mod is with a core of known low mu, a GDO, some patience and thorough temperature testing. (And preferably a speccy also).

One final thing I will point out. It is interesting to watch a spectrum analyser as the inductance is varied in this stage...... Remember, there are numerous iterations of the 847, and early models in particular need observation with a spectrum analyser when dealing with 70Mhz TX.... Lets hope Ofcom don’t get too many complaints……

73 de Nige.

I think this is one good reason that the PA1O mod should be applied rather than this 'core insertion' one. Removing a capacitor is easier and more constant in it's results.
Posts: 42
Joined: Mon 3 Sep 2007, 09:34
Location and locator: JO01GM

Postby G6GVI » Sun 29 Jun 2008, 15:31

G0KSC writes:
Anyone trying to find a core to use on their FT847 in order to increase power, I urge you to read on!

For months I have been trying to locate a core suitable for my rig having read the excellent article by G4FUF. However, yesterday I stumbled upon the articles of PA1O showing a modification needing no additional components and having the same effect as the G4FUF mod.
However, with both his findings and my experience this was no detrimental effect on 6mtrs. Infact, the current draw on 6mtrs is now less than it was previously.

Basically, the mod requires the simple removal of a surface mount capacitor from the bottom of the PA board. Sound to good to be true? it did to me but having not been able to locate an appropriate core, I gave it a go.

MY Ft847 now puts out 85 watts on 4 mtrs with just 17.1 amps draw and 100 watts on 6mtrs for 16.1 amps.

I thought it worth pointing this article out because whilst it is on this website, the article name does not lend itself to explaining the nature of this mod and as a result, is easily overlooked.

Posts: 136
Joined: Mon 24 May 2004, 18:02
Location and locator: Bolton IO83SN

Postby G8HVY » Mon 17 Nov 2008, 18:48

PA1O wrote

After my article on C5051 I opened the set again
to optimize L5006 just to find that the original value worked
best for me. The PA gain at 6m is (only) just sufficient, maybe
replacing C5051 with something like 100 pF would be a better
compromise and give 100W at 6m for everybody again at slightly higher current (and lower power at 4m)... Anyway, I'm happy with the present

So for max gain at 70 MHz indeed keep using the original value for L5006 as you already found out.

It should be possible to get 100W at 70 MHz with <23A current, but we are still lacking gain in the driver stage. Does anybody feel challenged? I'm rebuilding my shack at the moment, so I have no time to try and optimise further...

73 de Marc PA1O

Hi all

Since the follow up post from Marc PA1O I found that the best compromise
is to leave C5051 470 pf in circuit and remove C5050 390pf, as this is almost the same as fitting an additional 100pf in place of C5051
(recommended by PA1O) and it is a simpler solution.
The results that I achieved after removing C5050 were 6M 100w and 4M 80w output not bad.
Thanks to Marc for the original excellent article and his follow up comments.

73's Graham
Last edited by G8HVY on Mon 25 May 2009, 14:34, edited 1 time in total.
Posts: 63
Joined: Sun 20 Mar 2005, 19:44
Location and locator: IO90HW

FT847 Help

Postby G4FKI » Sun 30 Nov 2008, 14:50

I have just obtained a 847 with a serial starting 9125, which doesnt appear on any lists I have looked at so far.
Si Im curious as to the year of manufacture.
The CPU links are as follows:
1235 connected 46 open
The rig is tx/rx on 4m, power wise on full power its approx 25w, harmonics ie 2nd 3rd are around 60dbc or better but I have concerns of spurious signals around 40 to 50MHz.
It does not have the logic level switching transistor.

Any advice as to where I proceed would be appreciated.

Posts: 66
Joined: Wed 26 May 2004, 22:09
Location and locator: Ampthill Bedford IO92SA

Postby G8HVY » Sun 30 Nov 2008, 22:46

Hi David

If you check the Ser No again you will probably find that is is '9I25' which means that it was manufactured in July 1999 batch '25' If you still have the box you will be able to confirm whether the the set is a C2 version which is the UK version factory modified for 4m.


Posts: 63
Joined: Sun 20 Mar 2005, 19:44
Location and locator: IO90HW


Postby G4FKI » Sun 7 Dec 2008, 17:17

Yes it is 9I25, the box doesnt mention C2 at all, it has no switching transistor.
I have the covers of but cannot locate the TP 3011 etc to check for voltage , can anyone help with a pcb layout showing the testpoints.
Before I start delving further
I am concerned with unwanted Ip's before using the rig.

Posts: 66
Joined: Wed 26 May 2004, 22:09
Location and locator: Ampthill Bedford IO92SA

Postby G8HVY » Sun 7 Dec 2008, 18:15

The 'C2' region code can only be found on the white label on the box together with the serial number and other factory codes.
The Technical Supplement for the FT-847 can be found on Mods.dk web-site
under Yaesu manuals.
Best of luck.

Posts: 63
Joined: Sun 20 Mar 2005, 19:44
Location and locator: IO90HW


Postby G4FKI » Sun 7 Dec 2008, 19:07

Hi I found the yaesu tech supplement of another site but it didnt have a board layout ?, I will try mods.dk later in the week, as its no ,longer a free service.

Posts: 66
Joined: Wed 26 May 2004, 22:09
Location and locator: Ampthill Bedford IO92SA

FT847 Pa Board

Postby G4FKI » Sat 13 Dec 2008, 15:47

I have checked that the test point voltages are now all ok.
I am trying to work out the easiset way to gain access to the TX Pa board, it looks to be sandwiched in the middle, can anyone confirm the correct procedure to gain access to it ?

Posts: 66
Joined: Wed 26 May 2004, 22:09
Location and locator: Ampthill Bedford IO92SA

847 access

Postby G4FKI » Sat 20 Dec 2008, 16:02

Im still stuck trying to gain access to the PA/filter any advice appreciated asap.

Dave :?
Posts: 66
Joined: Wed 26 May 2004, 22:09
Location and locator: Ampthill Bedford IO92SA

Postby G5WQ » Tue 23 Dec 2008, 19:52

i have just finished modding my ft847 as per marcs PA1O information .

the capacitor i pulled out was the 390pf C5050 [ located closest to main pa transistors] , which leaves C5051 470pf in place .

i have actually ended up removing and playing around with different value capacitors in my ft847 and i find that it is best with a 470pf capacitor left in the board on my own particular radio .

i tried a 300pf cap in there but that resulted in 40w output .

i had to destroy both original capacitors in the end and have now fitted a new 470pf smd cap that i got from RS components , although i had to buy 25 as a minimum order, so if anyone needs a 470pf 500v SMD cap for their ft847 i have some spare .

to fit the new cap i superglued it to the pcb before soldering , works a treat !! .

i dont see 80w though, more like 70w , but with my psu at 13.8v .

if i raise the psu voltage to 14.5v the power out increases a fair amount , well thats obvious anyway .

my 847 is an 8H 0903?? version and made for uk market with C2 sticker on box .

i have the C2 version ALC cutback wire cut and no longer run the ferrite slug inductor mod .

i must admit that the spectrum analyser display is much cleaner now with marcs mod rather than the original spec , although i only have a basic spectrum analyser which runs to 1.5ghz .

i am seeing harmonics down at below -60db and the spectral display above and below the fundamental freq is much clearer than it was hi .

part number for the 470pf smd caps is 652-0881 from rsww [RS components ], although not cheap considering having to buy 25 at a time @ 40pence each .
they are smaller width than the original smd caps but still fit across the solder pads without problem and are easy to solder so long as you have a lampstand magnifier etc .

i also note that you have to adjust the HF power adjustment trimmer on the filter board and not the 50mhz one , in order to alter 70mhz output etc .

for anyone who is going to mod their own ft847 please note that on the filter board there are 2x 2 pin multi connectors down in the rear corner of the board that will fit either socket , so make sure you know which came from where when reassembling radio .

to remove the actual PA board it is best to unscrew and desolder the SO239 sockets from the pins on the pa board before trying to remove the pcb from the chassis .
there are also 2 small earthing tags soldered next to the rf output pins , and you need to watch you dont knock these off when taking apart or reassembling .

to get at the PA board you remove the bottom cover from tcvr, then remove all connectors and power cables from the filter board .

take all screws out and then you can lift the filter board out in one piece .

under this is an alloy "sub chassis" with 4 screws holding it in place , then once thats removed there is another shield plate underneath which has a row of 3 screws up the centre and 1 central each side , mark where these screws go with a marker pen so you know which holes to put them in on reassembly , because some holes are used for the next plate installation above it .

the PA pcb has a couple of small coaxial connectors , 2 power cables and one multipin connector to remove , plus various screws for holding pa down and for holding the various transistors to chassis .

once the so239 connectors are unsoldered and removed, and all screws taken out etc you can just lift the pcb up and out .

marcs mod is to remove the smd capacitor closest to the output transistors , and this i what i did to my 847 at first and is the mod i prefer on mine , but i found that i had removed a 390pf capacitor and not 470pf .

now after trying 2 values of caps [305pf and 470pf ] and obtaining a new 470pf smd cap i have just the single 470pf cap in the position of C5051 c5050 and am seeing 70w+ output on 70mhz and 100w on 6m .

im not going to do any more on the radio now because i must have stripped and reassembled it 20 times in the last 3 days and i am at a point now where i think my own radio cannot be improved from the configuration the PA board is currently in .

i think mine is one of the earlier versions of 847 and so i dont know if there are differences between it and later models , but suffice to say that marcs mod works ok on an 8H version just in case anyone else is interested .

david G4FKI i have a few digital photos of the internals of radio when taking it apart if you want to see them , just mail me if you want me to send them to you .

73 ian


Posts: 19
Joined: Sun 8 Jul 2007, 21:27
Location and locator: south marston io91do

Postby G8HVY » Sat 27 Dec 2008, 18:11

Hi Ian

It looks like your findings agree with the advice I suggested in my earlier post on this subject on the 17-11-08.
I recommended that the removal of C5050 390pf is the best compromise to keep the 6M power output as close to 100w as in the original specification. Obviously there will be some differences between sets due to component tolerances and set variations but this mod is in general consistent on most versions of the FT-847 that I have come across.

73's and a HNY to all on this web-site.

Posts: 63
Joined: Sun 20 Mar 2005, 19:44
Location and locator: IO90HW


Postby G4FKI » Thu 8 Jan 2009, 23:46

I am looking to purchase the tx filter board, but cannot find any email details for G4FUF ?

Posts: 66
Joined: Wed 26 May 2004, 22:09
Location and locator: Ampthill Bedford IO92SA

Postby G0EHV » Fri 9 Jan 2009, 10:52


Try this one, it worked for me :D

Posts: 173
Joined: Wed 28 Jun 2006, 09:32
Location and locator: IO94FU

FT847 tx filter

Postby G4FKI » Sun 11 Jan 2009, 21:54

I managed to get in contact with G4FUF, unfortunately the tx filter is unavailable due to a lack of response for the item.

Posts: 66
Joined: Wed 26 May 2004, 22:09
Location and locator: Ampthill Bedford IO92SA


Return to Equipment

Who is online

Users browsing this forum: No registered users and 2 guests