The KA7OEI FT-817 pages

Extracts from correspondence

Front panel view of my FT-817
A Front-panel view of the FT-817 sitting atop the switching regulator.

As you might guess, I receive a fair amount of email in response to these web pages.  Most of it is SPAM (the price of placing an email address on the web...) but some of it actually has something to do with the '817 - and I try to respond to all of email in that category.  (If you sent email and I never responded, I'm sorry - it either got zapped by a SPAM filter owing to some unfortunate aspect of your message or ISP, or it got buried in the pile and was submerged.)  As you can imagine, I occasionally get email that is less than civil:  This is filed in the section called "Room Temperature I.Q. Club" (the link is at the bottom of the page - closer to where it really belongs...)

Although all of the (FT-817) email is important, occasionally a user will bring up something that I'll add to the web page(s) either as a correction or addendum.  Other times, a topic comes up that might not neatly fit onto a web page - but is worth bringing up as "something worth thinking about" - in the hopes that someone else will chime in with good ideas of their own.

Notice:  Attribution of comments to specific individuals is given only when they have given me explicit permission to do so.  Otherwise, identities are protected.

What follows are random bits and pieces on various topics.  They might make interesting reading... or not...  Some of these exchanges spanned several emails.  They have also been edited for length and relevancy as appropriate, but the salient points have hopefully been retained.

FT-817 Receiver dynamic range:

Q1:  "Hi, Very interesting site with much info and good ideas. One main question, has any one done any improvements to tackle the relatively poor receiver dynamics on an otherwise excellent QRP rig?"  (From Great Britain)

A1:  By "poor dynamics" I take it that you are referring to the way the AGC operates - rather than the absolute dynamic range?

One thing that annoys me about the '817 is the way the HF preamplifier is handled:  I'd much rather it default to "off" and that they call it a preamplifier rather than an "Intercept Point" device.

I haven't found the absolute dynamic range to be a problem - as I usually leave the preamp off on HF - but I've had others say that in Europe, the radio wilts in the presence of extremely strong 49, 40, and 31 meter SWBC signals.

My biggest gripe is the AGC - that it seems to "flatten" considerably above S9 or so.  This, to me, seems to have the effect of strong signals (those >S9)  having a faster AGC time constant than weak ones. This has to do with the db/volts ratio on the AGC, I suspect.

One thing that I'd like to try would be to add a circuit that would, in response the AGC voltage, change the bias through D1063 on the receive board (by adding current at the collector of Q1053 - on the main board in the IF section) such that lower current flowed through D1063 under strong signal conditions.  This would require the construction of a simple op-amp based circuit and it would have some "threshold" "gain" and "offset" settings on it - but I think that it could effect a significant improvement if properly implemented.  It would take some careful "microsurgery" to do this, however.

Am I close to the mark, or did I miss your point completely?

[reply]:  "I agree [re] agc, but in fact I was referring to rx dynamic range, you hit right on with the strong European station that produce cross mod on 160 80 and 40, it isn't quite so bad on the HF side of HF  i.e.. 30 and higher, though some strong signals locally do tend to ruffle the edges of any signal which does not happen with other rigs here!  The measured IM and CM blocking is not as good as the base station."

A2:  I have noticed that the [performance of the] '817 is a bit lower in that respect than many other radios - but then again, there are several things that I keep in mind:

- Turn the preamp off.  I'm happy enough to hear S-5 noise floor on 75 meters at night instead of an S9+ noise floor.
- Usually, signals on 80/75/40 are strong enough that using the attenuator is perfectly acceptable.
- Hey, it's a low-powered portable, tiny radio!

I haven't really poked around inside the '817 to determine where it's weak link on RX resides:  The preamplifier isn't likely (itself) to be overloading - but it sure doesn't help later stages.  Is it the MESFET mixer that is overloading?  Typically, FET commutating mixers like that are pretty good in terms of dynamic range.  Is it a later IF stage that is overloading?  The 2-pole monolithic crystal filter - while somewhat broad - should really prevent something >15 KHz away from ever causing a problem with the latter IF stages.

Not having experienced the sort of overload that you have likely experienced (I do, in fact, have a 50-100 kW shortwave station within 5 miles of my house - KTBN - that often operates on 7510 - but I've never noticed any overload, mainly because I'm well off to the side of their log-periodic's radiation pattern.)

Perhaps you can answer this question:  Does a strong signal on, say, 7500, impact reception on 7075 as much as it would on 7300?  That is, as long as it falls within the receiver's front-end filter, you have a problem?  If that is the case, then what I would guess to be the main problem would be the complete lack of  proper mixer termination at frequencies other than that of the [1st] IF.

In looking at the schematic, I see that T1019 - a quadrafilar-wound transformer - transforms the mixer's output impedance from something quite high (probably in the 200 ohm area) back down to something likely to be in the 30-50 ohms area.  As you can imagine, the mixer is only likely to be terminated properly at a frequency somewhere in the vicinity of the 68.33 MHz IF.  At other frequencies, the converted frequency is likely to be reflected back into the mixer:  A terrible thing to do if it is performance under high-signal conditions that you are after.

If this is the case, one might (ironically) enhance performance by adding another amplifier stage - between the mixer and the 1st IF transformer (between T1019 and T1023, specifically.)  This might take the form of a JFET grounded-gate amplifier or a decent MMIC - something that provides reasonable termination over a wide range of frequencies. Also, a simple L/C/R diplexer arrangement would also be helpful (to keep too wide a frequency range from needing to be terminated properly.)  Perhaps following this amplifier with an attenuator (to maintain the receiver gain) would be appropriate.

Also, a simple 3db [or so] resistive pad may also help things tremendously: This would increase the return loss (to the mixer) by 6 db and could improve IMD performance by much more than that.  (I kind of doubt that the "reduce signals by 1db to effect 3db decrease in IMD" rule could be extrapolated that far - but it might help.)  3db of extra loss at this point would probably not be noticed - as the radio has enough conversion gain anyway.  I will be the first to admit that I don't have a good "feel" as to how reflections on the mixer ports of MESFET mixers affect their performance - but I suspect that it is similar to that of diode-ring DBMs.

One radio that does this is my old Drake TR-7:  It's receiver is much like the R-7 (in the front end, anyway.)  From the antenna port, it goes through a Low-pass filter, then a high-pass, and then into a high-level diode-ring mixer.  From there, it is diplexed (I think) and goes to a grounded-gate amplifier using a U310 FET - and then goes through a 4-pole filter at 48.75 MHz (I believe.)  This radio isn't quite as sensitive as most (about 0.5 microvolt) but I have always been able to hear more noise on my 10 meter antennas than on my dummy loads - and I never had a problem with it overloading - even with another amateur down the street operating on the same band.  (He was on 40 CW down below 7.1 MHz and I was on 40 phone - at about 7.2 MHz or so) aside from some key clicks - that the noise blanker took out.  (I might have bothered his receiver, though...)

While on the subject of "bad receiver design" (the '817 isn't what I'd call "bad" - but it certainly could use a bit of improvement...) I did some work to improve some Kenwood radios of my own and of some other friends.

The radios in question are the Kenwood TM-732 and TM-733.  These are [typical] dual-banders of early '90s vintage - and internally, are very similar in design.  The '733 added an "IPO" circuit that supposedly improved dynamic range - but it did that by changing the gate bias on the 2nd RF amplifier's dual-gate FET.  (Yes, it had at least TWO RF amplifiers preceding the mixer.)

The problem with the '733 was that (on VHF, anyway) there was an RF preamplifier (dual-gate FET) followed by another RF amplifier (a nearly identical dual-gate FET) followed by the mixer - which was also a dual-gate FET.  As you can imagine, with 30-40 db of gain in front of a mixer that itself had 10-15 db of conversion gain, you could expect the radio[s] to overload badly around town - and they do!

The specifications for the receiver are 0.16 microvolts for 12db SINAD (as I recall) and this receiver came in at about 0.1 uV or so.  Even the 0.16 uV rating is rather silly, as the vehicle environment is likely to present a noise floor (due to vehicle electrical noise and/or being in "noisy" areas around town) of a comparable value.

I decided to keep the radio "within specs" however, and I ended up adding about 12db of resistive attenuation between the 1st RF amplifier and the 2nd RF amplifier - and that resulted in radio's sensitivity staying in the 0.15-0.16 uV area and a "theoretical" increase in IMD resistance by 36db.  (The amount of attenuation was measured by doing "before" and "after" measurements of signal levels required to obtain a given "S-meter" reading.)  The result?  The radio actually does a pretty good job around town now and doesn't get overloaded nearly as often.  On 70cm, the RF stages were different - and it only took about 3-6db (I don't recall exactly how much) to put it "just" within specifications and it is still somewhat susceptible - but...  (Modification info for the TM-733 may be found here.)


Other topics will be added in the future...

Work continues on these pages - please revisit soon!

Important Notice:  The information contained on this and related pages is believed to be accurate, but no guarantees are expressed or implied.  The information on this and related pages should be considered to be "as-is" and the user is completely responsible for the way this information is used.  If you find information that you believe to be incorrect, missing, or should not be here, please report it (in detail) via email.  Anonymous or vaguely-worded reports will likely be ignored.

Consistent with human nature, not all email in response to this page is civil:  A very small percentage of it could be considered to be rude, at the very least.  For the morbidly curious, there is the Room Temperature I.Q. Club page.  Warning:  Although harsh language has been softened, it is still possible to be offended by what is there...

Go to The KA7OEI FT-817 "Front Page" - This is, well, the "front" page of the '817 pages here...

Any comments or questions?  Send an email!

This page maintained by Clint Turner, KA7OEI and was last updated on 20021030.  (Copyright 2001-2002 by Clint Turner)

Since May 15, 2001, this page has been looked at at least this many times: 

Since 12/2010: