compiled by William Eric McFadden

Each author retains copyright.

From Ed Manuel, N5EM, on 04 Sep 1995

I got mine after waiting about 4 months. Of course, I bought mine with the great QRP-L group purchase of 37 units. I'm sure that Index was glad to have the order but it severely taxed their production. It was worth the wait. I am sure that I have not used mine as much as others, but here are my observations.

You will not be disappointed in your observations. It is a miracle of packaging. First of all, this puppy is small. Well, maybe not like a NorCal 40A, but then it is general coverage 1.8 to 30 mhz. with all ham bands in between! The functionallity from the front panel is nothing short of amazing. With just a few buttons (which you have to work with a bit to learn), you can change frequency, use/load a memory, change the bandwidth from 100 hz. to 2400 hz. in 100 hz. steps, change the keying speed (of course the keyer is built-in) in increments of 5 wpm, change mode from SSB to CW, change the tuning rate from slow to fast or go to RIT / Split mode. Keying is very nice from on-air reports. Hope to get some SSB use in pretty soon, but I don't operate much SSB. Planing a couple of Cub Scout campouts this October where we'll give the SSB mode a real workout.

Now, a couple of wishes.

I wish that the power control was on the front panel. You need to adjust it when moving from band to band. If, for example you are doing work at .95 watts, you will have to readjust when you change bands. I realize that it would be an engineering challenge to move it.

I wish that the keyer speed was incremented in 1 wpm steps. 5 wpm is acceptable but not best. I wish that the radio had the ability to load at least one of the memories with different frequencies. I would like to work RS-12, mode K (21 mhz. TX, 29 mhz. RX). The radio will do it (I think) with the RIT but imagine cranking the freq. from 21 to 29 mhz. in 100 hz. steps. I realize that notmany folks would want to cross band, but I think that a small software change might get this one and allow RS12 satellite operation.

I wish the radio had an accessory cover/case for portable operation. Specifically, I'm thinking about a padded leather case to keep me from scratching it up when I take it out to a Cub Scout outing. I will eventually make one but I'd have gladly paid $40 for one designed for the radio. As a note, I always spend the $20 or so to buy a soft cover to protect my handheld radio. In fact, now a days, you have to buy two, one for the short battery and one for the long battery. So I will end up spending $40 to protect my $500 handheld radio. Why not for my $600 HF radio?

I wish that I had a VOX. I can live without it. I will probably end up makeing one of these outboard. But, it's better for VOX to be integrated to get smooth anti-vox. Why VOX? Makes headset operation so much nicer, especially if you are using headphones anyway. I realize that there are probably good engineering reasons for not having a VOX. This is my wish list, remember? It would cost more, too. At $600 it's right at the top end of the target market, I suspect. Another $50 to $100 would dramatically reduce its marketability. Maybe in the dreamed-of, matching QRP Plus accessory? (Please, don't call Bruce and ask about this. This is my wish list, remember :-)

I wish the LCD display had a backlight for night operation. It would need to be switched on and off. Maybe from combinations from the front panel in software?

Now, negatives.

There are a few birdies. I haven't tried to find them all. Found one at about 21.4 mhz. None on 40 or 30 that I have noticed. There is the documented W1AW problem. I haven't even followed this one enough to be able to properly describe it. I don't hear W1AW that well.

The radio has a documented susceptability to high power transmitters close by. If you buy a Plus, don't even hook it up if a regular 50-100 watt radio is near by. I wish I could define "near by". Certainly not in your shack! It'll smoke the finals and a few other parts for good measure. Maybe not within a few houses of yours. Don't know here. And, CERTAINLY, not to a field day type outing where high power rigs are on simultaneously. For me, I operate QRP exclusively. I am concerned about my satellite/ATV station. I have a 100 watt 70 cm amplifier for ATV and sats. For now, there is NO CHANCE that the Plus will be hooked up (any wire) when that amplifier is operational.

In fairness to the Plus, I really don't know how many folks have had this problem. It may be a precious few. But, if I had it, you can be sure that my displeasure would be vocal too.

Would I do it again if I had the chance? ABSOLUTELY. It, like every other radio I have ever owned, has its idiosynchrosies (bet I blew that one). Part of the character of any radio is its uniqueness. I, like everyone else, will not tolerate fundamental design flaws. I don't think anything I have discussed, except the high-power susceptability, is a fundamental design flaw. I expect that this one will be fixed. I will probably send mine back for upgrade when it is available. But I love the radio.

If you ever operated one of AEA's DX handies (10m/6m) walkie-talkie size QRP rig, or one of the Mizuho single banders, you know what I mean. They are a speciality rig and you adjust to the limitations. I don't mean that you learn to think of the rig as perfect. Like people, none of them are. But, the Plus fits me very well. I am almost 50 (and may decide to stay there for years to come) and I've made some mistakes with rigs in the past. I sold my Icom 502. I regret it to this day. I sold my TS-130V. Only the compassion of the fellow I sold it to allowed me to reacquire it. I will never let it go again. The Plus will probably be like that. Part of the estate when I finally go QRT.

There you have it. My, not entirely dispassionate review of mine.

From Larry East, W1HUE, on 07 May 1996

I was able to borrow a New and Improved QRP-PLUS (hereafter referred to as "QRP++") from a local ham (Rob, WS7U) last weekend and ran some comparison tests with my modified "original" QRP-PLUS (hereafter referred to as "QRP+"). I was also able to borrow an IFR-1500 service monitor to use in getting data on spurious emissions, receiver sensitivity, etc.

I have made a lot of modifications to my QRP+, so some of the comparisons may not apply to a "stock" QRP+. The mods I have made include:

Additional grounding and bypassing in an attempt to reduce internal receiver noise (not sure how successful this has been...)
Added 22 pF cap across middle inductor in the first mixer LP filter for better rejection between about 32 and 70 MHz.
Made the buffer amp between the xtal filter and mixer (transmit path) resonant at 50 MHz to reduce xmitter spurious emissions.
Replaced the transmitter LP filters used for 30M through 10M with elliptical filters to improve harmonic rejection.
Added an audio limiter to reduce effect of SCAF (Switched Capacitance Audio Filter) delay on the AGC (infamous "AGC thump problem" on strong signals).

Other mods have been made for "operational convenience" (such as an LED to indicate when SHIFT or RIT is on, etc.) which should not affect actual performance.

I am writing up all of these mods and they will be published in The QRP Quarterly and available from the QRP-L FTP site and/or WWW home page (one of these days...)

Anyway, on with the comparison with my "Super QRP+" and the "QRP++" (serial number 1353).

Receiver

My general impression is that I do NOT like the receiver on the QRP++ as well as my QRP+. In my opinion, it has WAY too much IF gain which compounds the "AGC thump" and strong adjacent signal leak-trough on narrow settings of the audio filter. However, it is somewhat more sensitive, has less "internal noise" and can do a little better job of pulling in very weak signals on a quiet band. On the other hand, I was able to copy weak SSB signals on a noisy band (75M) slightly better on my QRP+ -- probably due, in retrospect, to my audio limiter clipping the noise peaks.

Anyone who claims that the QRP++ does not have an AGC thump problem on strong signals has either been listening to a dead band or using a rubber duck for an antenna! It is as bad or worse as in a QRP+ (without the audio limiter -- I disabled it for comparison with the QRP++). It is not as severe when the SCAF is at its widest setting, but it is certainly there. The high IF gain of the QRP++ makes the thump much worse on strong signals and also makes the 20dB input attenuator less effective in reducing the thump.

Adjacent strong signal interference when the SCAF is narrowed up for CW operation is worse in the QRP++ than in my QRP+. This an "aliasing" problem in the SCAF and NOT a problem with the receiver front end or IF. The higher IF gain and higher LP filter roll-off in the QRP++ probably both contribute to the problem.

Receiver sensitivity: I attempted to determine the "minimum detectable signal" using the calibrated RF generator in the IFR-1500. By "minimum detectable", I mean a signal that can be clearly heard above receiver noise and could be copied if CW (probably could not be copied if SSB). My somewhat subjective results using a filter width of 0.5 kHz are as follows:

QRP+ (modified) -- Between 0.15 uV and 0.3 uV, depending on band.
Best on 160, 12 and 10M (0.15uV) and worst on 40, 30 and 20M (0.3uV).
About 0.2uV on the other bands.

QRP++ -- 0.1 uV or slightly better on all bands.

I could definitely copy very weak CW signals on 20M better on the QRP++.

S-meter: The S-meter on the QRP++ acted very strangely; it would register zero on all receiver signals below about 15uV and then suddenly jump to S-5 or S-6. This was not a problem with the AGC; output audio level remained very constant for all signal levels above about 1uV. Maybe a malfunction in this particular unit? (The owner balked at me removing boards from his new "pride and joy"...) S-meter sensitivity is also too high in my opinion: S-9 corresponded to input signal levels of 45 to 50 uV on all bands except 80 where it was only 32 uV. On the other hand, my QRP+ requires an input signal level of between 200 uV (160M) and 250 uV (30M on up) for an S-9 reading.

Transmitter

As I am sure you are all aware by now, the QRP++ includes an internal speech processor for "improved" SSB performance (depending on whether or not you like a lot of "audio punch" with that definite "processed" quality). Mike gain and compression level are automatic and cannot be changed without changing component values. The speech processor makes use of an op-amp with an FET in the feedback loop to control the gain; it is similar, in concept at least, to the $9.00 speech processor kit sold by Ten-Tec. (That's got to be one of the best buys around!). The Ten-Tec kit, by the way, should work well with the QRP+ -- but as an external, NOT internal add-on; I doubt that you could find room for it inside the already pretty full QRP+ box. I did not make any SSB tests on the QRP++ for two reasons: my main interest is CW operation and I did not have the time.

The QRP++ uses an ALC (automatic level control, not be confused with the receiver AGC, automatic gain control) to control the output power on CW as well as SSB. (The QRP+ uses ALC on SSB only). This has the advantage of keeping the output level quite constant from band to band, unlike the QRP+ which can change by as much as 2 or 3 watts between bands. The maximum output power of the QRP++ I tested was about 5.5W on the low bands, dropping to about 5W on 10M.

Keying: First the good news -- the QRP++ can follow an external keyer to at least 42 WPM, whereas an external keyer can't be used above about 28 WPM with the QRP+ (no biggie for me... :-) The bad news is that the keying on the QRP++ is very "hard"; the transmitter output waveform has a rise time of about 0.5ms and a fall time of less than 0.25ms! However, it did not sound as bad as expected when monitored on another receiver, but there was a definite key-up click. My QRP+, on the other hand, has rise and fall times of about 2.5 ms at 5W out and the keying sounds very clean (none of my mods should have affected keying characteristics.) My QRP+ output rise-time shortens up somewhat at higher power levels, but still sounds clean.

Harmonic output: The FCC requires that spurious output from amateur transmitters operating below 30 MHz be at least 30dB below the carrier for 5W or less carrier output and at least 40dB for greater than 5W output. The three transmitter LP filters used for 30 thru 10 meters in the QRP+ are barely adequate to provide 30dB second harmonic attenuation on 30, 17 and 12M. Index claims that the filters are "quite adequate" and did not change the design for the QRP++. They are perhaps adequate for the QRP++ which (without modifying the ALC circuit) has a maximum output of about 5W. However, most (unmodified) QRP+'s are capable of 7 - 8W out on bands where the harmonic attenuation is likely to be less than 40dB...

Harmonic output measurements made with an IFR-1500's spectrum analyzer are given in the following table. The "Std QRP+" column lists measurements made on my QRP+ before filter modifications, and "Mod QRP+" is after the new filters were installed for 30 through 10M. All measurements were made at 5W carrier output.

	             |    Harmonic output in dB below Fo
	    Carrier  |   QRP++    |  Std QRP+  |  Mod QRP+
	   Freq. (Fo)| 2xFo  3xFo | 2xFo  3xFo | 2xFo  3xFo
	  -----------|------------|------------|------------
	     1.820   |  48   >60  |  50   >60  |  52   >60 
	     3.560   |  55   >60  |  55   >60  |  55   >60
	     7.040   |  55   >60  |  52   >60  |  55   >60
	    10.116   |  31    52  |  34    53  |  50   >60
	    14.060   |  45   >60  |  50   >60  |  55   >60
	    18.070   |  32    57  |  40   >60  |  53   >60
	    21.060   |  38   >60  |  37   >60  |  48   >60
	    24.900   |  33   >60  |  36    54  |  58   >60
	    28.060   |  38   >60  |  48   >60  |  58   >60
	

Differences between the QRP++ and standard QRP+ are due to component tolerances and relative measurement errors. I estimate my relative measurement errors to be on the order of 3dB.

Spurious emissions other than harmonics: An unmodified QRP+ has several "spurs" that are less than 30dB below the carrier on all bands from 20M on up, but no spurs within 40dB of the carrier on lower bands. The spurs on my modified QRP+ are all more than 40dB down -- except for 12M (see below). All spurs from the QRP++ were at least 40dB down except for one about 6 MHz below the operating frequency on 10M that was 30dB down, and 12M where spurs were 32 to 36dB down depending on frequency (see below).

12M has a special problem: Mixing products with the second harmonic of the 50 MHz IF can result in spurs very close to the 12M carrier frequency! The mixing products that are of concern are given by:

	     Fo +/- N x(2xIF - (LO + Fo))
	where:

	     Fo = carrier frequency = LO - IF,
	     IF = 50 MHz,
	     LO = Local Oscillator frequency,
	      N = 1, 2, 3, ...
	

For example, at a carrier frequency of 24.9 MHz, mixing products can cause spurs at +/- 200 kHz, +/- 400 kHz, etc. from the carrier (decreasing in amplitude with increasing N).

In my QRP+ running 5W out, the "first order" (N = 1) spurs due to this mixing are quite strong; only 28dB down at Fo = 24.900 MHz and 15dB down at Fo = 24.980 MHz (where the spurs are only 40 kHz from the carrier)! If the output is reduced to 3W, these spurs are 34dB and 20dB below the carrier, respectively. It appears that the strength of these mixing products is related in a very non-linear way to the "carrier injection" level (i.e., the IF signal level). Thus when the carrier injection is reduced to give an output of only 3W, the mixing products are drastically reduced.

In the QRP++ I tested, the situation is much better; the spurs are down 36dB at Fo = 24.900 MHz and 32dB at Fo = 24.980 MHz (just barely below the "legal limit", but below never-the-less). This improvement is due, apparently, to added gain in the transmitter RF chain resulting in less carrier injection into the mixer for a given transmitter output. An MC1350 amplifier IC is used as the "pre-driver" in the QRP++ and a two stage desecrate amplifier is used in the QRP+. A different mixer is also used in the QRP++ which might also affect the level of the mixing products.

I don't know if my QRP+ is typical or not. I have asked for 12M measurement results from other owners but none have been forth coming; perhaps other QRP+ owners are embarrassed to tell about what they are seeing!

It appears that adding some gain in the QRP+ transmitter would improve the situation. Maybe I can come up with a way to do this without requiring a major redesign effort; in the mean time, I'll keep the power level at 3W or below on 12M and operate below 24.950 MHz -- if, indeed, I operate on 12M at all!

Nit-Picking Stuff

I don't like the QRP++ S-meter; besides being too sensitive it is harder to read than the one on my rig.

The QRP++ has a slightly smaller tuning knob; I like the old one better.

The QRP++ is supposed to have a varistor to protect the receiver mixer. However, in the rig I tested, there were pads on the RF board (the top one -- I only had to remove the cover to see it Rob!) marked "VR1" but nothing was there. Kinda makes one wonder...

Summary

Good points about the QRP++ relative to an unmodified QRP+:

• It has less receiver noise and greater receiver sensitivity.
• Power output is relatively constant from band to band.
• More "punch" on SSB -- or so I've been told (I haven't heard one on the air).
• Transmitter output is cleaner (but it still suffers from marginal harmonic reduction on some bands).

Things I don't like about the QRP++:

Its receiver IF gain is too high.

"Thumping" on strong signals is very severe (could be improved by adding an audio limiter after the filter).

Strong signal "bleed thru" at narrow filter settings is worse than with a QRP+.

Keying is too "hard".

Final notes:

It is not my intention to "knock" Index Labs rigs; on the whole they are nice little boxes and pretty much of a bargain compared to other rigs available. However, there can always be improvements -- and Index should REALLY be embarrassed about the 12M spurs!

Judging from comments I have seen on QRP-L, it appears that the QRP++ configuration is in a bit of a flux; problems one owner notes don't seem to be seen by other owners. Hopefully Index will soon get the production kinks worked out and start providing a "reproducible" product.

Am I planning to upgrade my QRP+? Not on your life; I have put too much time and effort into modifications and I don't want to start all over again! Except for the 12M spurs -- which I WILL cure someday -- I like my little rig just fine!

From Larry East, W1HUE, on 18 Sep 1996

I finally got around to installing the "Rev 4C" EPROM from Index in my QRP-PLUS. This is a "supplied upon request" $6 upgrade for the original QRP+ (not the new improved version) that is intended to cure a couple of keying problems. In particular, it eliminates the "forced spacing" of the internal keyer that so many folks objected to and it allows an external keyer to be used up to at least 40 wpm. Turns out that characteristics of the SCAF (switched capacitance audio filter) are also changed by this upgrade. The EPROM that I replaced was "Rev 03"; that's the one that changes the internal keyer to provide 1 wpm rather than 5 wpm speed increments and "remembers" the last operating mode (CW or SSB). The Rev 03 new features are still present in the Rev 4C code. I decided to keep the Rev 03 EPROM (I sent Index $10 rather than $6 and the old chip) for a "backup" and so that I could make some direct comparisons with the Rev 4C code. I present the results of my comparisons here for your information and edification.

Keying Characteristics

The Rev 4C code changes did indeed eliminate the "forced spacing" characteristic of the internal keyer. The "feel" of the internal keyer is now the same (or very close to) "Curtis B mode" keyers and is much easier (for me) to use. The rig will now follow an external keyer up to at least 44 wpm (that's as fast as my keyer will run), whereas the Rev 03 and earlier EPROMs would not allow external keying above 27 wpm or so. However, keying becomes erratic above 42 wpm when in RIT or SPLIT mode with either the internal keyer or external keying, so for all practical purposes the keying speed is limited to about 40 wpm.

A disappointing side-effect of the Rev 4C code is that keying is very mushy when in normal XCEVE mode -- the rise time of the transmitter output envelope is almost 10ms! When the front panel mode switch is in the RIT or SPLIT position, the keyed output risetime is about 2ms -- the same as with the Rev 03 code -- resulting in much better sounding keying. "What's going on here," you may well ask (as did I). To answer that requires a little explanation of the keying sequence in the rig. When a key closure occurs, +12V is immediately applied to the transmitter driver stages and the antenna is switched from the receive to the transmit RF chain. However, the actual keying signal (a voltage applied to the balanced modulator to unbalance it and let RF pass) is delayed slightly -- about 5ms in XCEVE and about 25 ms if RIT or SPLIT is turned on. (This delay was much longer with previous firmware versions which resulted in the limited external keying speed.) The short delay when in XCEVE does not allow the bias on the first driver transistor to reach its final operating point before RF is applied resulting in the slow rise of the output RF envelope.

So -- does that mean that one must leave the mode switch in RIT or SPLIT to get decent keying? Well, not if you are willing to make a simple modification to the RF board: decrease the size of C54 by a factor of 10 from 4.7uF to 0.47uF. That simple change will result in a 2ms risetime when in XCEVE as well as RIT and SPLIT modes. The RF board is the top one in the stack, and C54 is located in the upper left-hand quadrant (when viewed from the front of the rig).

Another problem noted with the Rev 4C EPROM: an RIT or SPLIT receive/transmit offset of more than 3 to 5 kHz (depending on the band) resulted in noticeable "chirp" on the keyed signal monitored on another receiver. The chirp became very pronounced at offsets of 10kHz or more. With the Rev 03 EPROM installed, I could not detect any noticeable chip below a receive/transmit offset of 15 to 20 kHz (depending on band). The Rev 4C EPROM provides less delay between key-down and transmitter output in order to accommodate faster external keying (see above), but this reduced delay does not give the PLL sufficient time to stabilize the VCO frequency. I have seen complaints from others about chirp with this EPROM even with no RIT offset, so the PLL settling time may vary somewhat from rig to rig.

The good news is that the chirp CAN be cured -- at least for receive/transmit offsets that are not too great -- by replacing a couple of caps on the LO board. The bad news is that the LO board is the third one down in the stack. I changed C43 and C44 from 4.7uF (original values) to 2.2 uF, and now I don't notice any keying chirp for receive/transmit offsets of at least 20 kHz. This is good enough for "DX split" operation, which is typically no more than 10 kHz on CW. Some care is in order when mucking about with these cap values; reducing the values too much can result in the PLL becoming unstable. I did not try to determine how much lower than 2.2uF can be safely used since that value seems to do the trick. C43 and C44 are located near the LO board's right-hand rear edge (when viewed from the front of the rig).

I have seen reports that the Rev 4C EPROM produces more pronounced key- clicks in the side tone. I have installed mods in my rig that greatly reduce keying clicks and thumps in the audio output, and I did not notice any change in the sidetone after installing the new EPROM. However, the reduced keying delay (with the new EPROM) could well result in a noticeable increase in the key-down click in unmodified rigs. (The descriptions of these modes were included in my posting for reducing AGC thump some time ago and are available on request.)

SCAF Response Changes

Replacing the Rev 03 EPROM with Rev 4C chip resulted in different SCAF bandpass characteristics for filter width settings of 1.0 and below. The approximate 6 dB bandpass edges and center frequencies for a few filter settings are compared in the table below (frequencies are in Hertz).

     Width   |            Rev 03          |             Rev 4C
    Setting  |    low     center   high   |     low     center   high
  -----------|----------------------------|----------------------------
      0.8    |    410      905     1400   |     600      1095    1590
      0.4    |    410      705     1000   |     600       895    1190
      0.2    |    410      600      790   |     600       795     990

My filter width measurements weren't the most precise, but they WERE performed with the AGC disabled (just in case anyone asks...).

The new filter characteristics are more consistent with the receive/transmit offset of about 700 Hz (I measured 720 Hz on my rig). Using the narrowest filter settings (0.3 and below) with the older EPROMs can result in inadvertently listening about 100 Hz (or more) below the transmit frequency if the received signal is centered in the filter passband.

Copyright (c) 1996 by Larry East, W1HUE
May be reproduced if proper credit is given and not quoted out of context.

From Larry East, W1HUE, on 09 Sep, 1996

A point of clarification -- unless the recent review in QST prompted Index to make a change, the transmitter output filters in the "New Improved QRP++" are the same as the ones in the "Old QRP+". The filters for 160M thru 40M are "elliptic" design and provide at least 40dB of harmonic attenuation. The filters for bands above 40M are shared between two bands -- one filter for 30M and 20M, one for 17M and 15M and one for 12M and 10M. These filters are NOT elliptic (for some strange reason) and in some cases the harmonic attenuation is less than 40dB but (usually) greater than 30dB. The US FCC requires spurious radiations to be at least 30dB below the carrier for amateur HF transmitters running 5W or less, and at least 40dB below the carrier above 5W. A QRP+/QRP++ will (usually) meet this requirement if output power is limited to 5W, but not on some of the higher bands at >5W output. Some countries require harmonic emissions to be >40dB below the carrier at any output power level.

My previous post stated that I have new output filter designs that will provide >45dB harmonic attenuation for the bands above 40M. (Send me email if you want the design info.)

There is another problem regarding 12M spurs that CANNOT be eliminated by the output filter as they are very close to the carrier frequency. These spurs are (just barely) >30dB below the carrier in the QRP++, but NOT the QRP+ (at least not mine). I have an idea for fixing the 12M spur problem, but I haven't had tome to try it. Hopefully I'll have it done before the sun spots return. :-)

I hope that everything is all perfectly clear now... :-) :-)