There does seem to be some variability. With the MAR-6, I was getting -15dBm out, but I had to keep the level on each modulator input pin down to 135mv pk-pk to avoid IMD products raising above the noise floor.
I've changed the MAR-6 MMIC for an MSA-0886 and I can now increase the pk-pk to 295mv, giving -4dBm out. I've also put an inductor in the supply to the MMIC as close to the MMIC as possible (track from L5). The only snag with the MSA-0886 is that it needs 10v.
I'm still trying various things, but my access to an analyser is limited.
Brian
power from digilite on 23cm
Re: power from digilite on 23cm
Hi Brian G4EWJ,
Thank you for your anwser. Yes, that makes sense to me that you could not acheive the expected level of modulator power output without creating too much IMD products (spectral regrowth) distortion in the final board RF output.
Do you think the low power issue could also be related to a companion thread in this forum "when is a MAR-6 not a MAR-6"??
Thanks again...de Ken W6HHC
Thank you for your anwser. Yes, that makes sense to me that you could not acheive the expected level of modulator power output without creating too much IMD products (spectral regrowth) distortion in the final board RF output.
Do you think the low power issue could also be related to a companion thread in this forum "when is a MAR-6 not a MAR-6"??
Thanks again...de Ken W6HHC
Re: power from digilite on 23cm
Please let us know how it goes with the MSA-0885. It's the MMIC I'm planning on using. I was hoping to have this done by now, but on my week off.. the XYL had other plans and I've been finishing a room in the basement for the kids instead. 
I decided to go with the MSA-0885 because they appear to still be readily available. I'm wondering if two stages of ERA type MMICS might be better still.. We will need to experiment with it some.. when I get back to working on it.
Fred W0FMS
I decided to go with the MSA-0885 because they appear to still be readily available. I'm wondering if two stages of ERA type MMICS might be better still.. We will need to experiment with it some.. when I get back to working on it.
Fred W0FMS
Re: power from digilite on 23cm
The MSA-0885 seems to be a straight pin version of the MSA-0886. I've changed the regulator to 10v and it seems to work fine. It must be better to have a device with a higher P1 point than the MAR-6, but I haven't tried a real MAR-6 yet.
>>Do you think the low power issue could also be related to a companion thread in this forum
>>"when is a MAR-6 not a MAR-6"??
Definitely! It explains why I didn't get much gain after fitting it and why I had to keep the IQ levels into the modulator very low to avoid regrowth.
Brian
>>Do you think the low power issue could also be related to a companion thread in this forum
>>"when is a MAR-6 not a MAR-6"??
Definitely! It explains why I didn't get much gain after fitting it and why I had to keep the IQ levels into the modulator very low to avoid regrowth.
Brian
Re: power from digilite on 23cm
I have been looking at power from my DigiLite board and this has raised a question:-
How are users measuring the power out? A power meter or a spectrum analyser? With modulation or with carrier only (no mod)? Each method gives a very different result.
In my case on a spectrum analyser the unmodulated carrier is over 10dB higher than the modulated envelope.
Dick G4BBH
How are users measuring the power out? A power meter or a spectrum analyser? With modulation or with carrier only (no mod)? Each method gives a very different result.
In my case on a spectrum analyser the unmodulated carrier is over 10dB higher than the modulated envelope.
Dick G4BBH
Re: power from digilite on 23cm
Getting the same result Dick; on 23cm about -4dBm with mod and +5dm without using a MSA-0886 for the MMIC and a 10v regulator.
On 70cm using the MSA-0886 I've had to increase the pad between the AD8345 and the MMIC to 12dB and get 0dBm with mod and +8dBm without.
Using a power meter will measure the wanted and harmonic / spurious which are considerable; my HP 432 power meter with its 478A head measures around 6dB more than the unmodulated carrier on the SA.
I think Brian G4EWJ's mod with the 10v regulator and the MSA-0886 should be made into an official upgrade as the P1dB on that MMIC is around 10dB more than a (genuine!) MAR-6. I run 900mV pk-pk of baseband into the modulator. Clive G3GJA.
On 70cm using the MSA-0886 I've had to increase the pad between the AD8345 and the MMIC to 12dB and get 0dBm with mod and +8dBm without.
Using a power meter will measure the wanted and harmonic / spurious which are considerable; my HP 432 power meter with its 478A head measures around 6dB more than the unmodulated carrier on the SA.
I think Brian G4EWJ's mod with the 10v regulator and the MSA-0886 should be made into an official upgrade as the P1dB on that MMIC is around 10dB more than a (genuine!) MAR-6. I run 900mV pk-pk of baseband into the modulator. Clive G3GJA.
Last edited by G3GJA on Sat Feb 04, 2012 3:30 pm, edited 1 time in total.
Re: power from digilite on 23cm
I reckon the only consistent way is to put it into LSB or USB test mode to give a single carrier output, which should be read the same by all power meters and call this the PEP. This can then be related to the capability of amplifiers further down the chain, particularly the P1 point.
The average power from the DigiLite will vary from unit to unit according to the tolerance of the components in the Nyquist IQ filters. The spectrum analyser display with normal modulation will move up and down according to the resolution bandwidth in use, so that's not an absolute either. Different meters will react to amplitude modulated signals differently. It's not very useful with SSB to use the power reading when 'aar-lowing' into the mic. Much better to whistle it up to maximum and use that figure as the PEP.
Brian
The average power from the DigiLite will vary from unit to unit according to the tolerance of the components in the Nyquist IQ filters. The spectrum analyser display with normal modulation will move up and down according to the resolution bandwidth in use, so that's not an absolute either. Different meters will react to amplitude modulated signals differently. It's not very useful with SSB to use the power reading when 'aar-lowing' into the mic. Much better to whistle it up to maximum and use that figure as the PEP.
Brian
Re: power from digilite on 23cm
Clive
With regard to your comments re the MSA-0886 I agree this could be an advantage. However the 10V regulator has some some drawbacks. Firstly the extra 2V on the 5V regulator will increase it's dissipation so it will get much hotter. Also most 10V regulators have a voltage drop of 2V or more. At present the DigiLite board will work happily down to 11V but increasing U2 to a 10V regulator could mean problems with a supply of less than about 12.5V which might not be so good for portable use. A three pin LDO 10V regulator is the answer but is hard to find in an M version. The LM2940T-10.0 is probably suitable. Also the Ultram oscillator will require a separate regulator as it will run happily on 8V but 10V is a little over the top.
Just finished a 24 hour soak test of my DigiLite board with no obvious problems except for a vast video file created by GBPVR
Pity there are no DATV stations near Dover and it's too cold to go portable!
Dick G4BBH
With regard to your comments re the MSA-0886 I agree this could be an advantage. However the 10V regulator has some some drawbacks. Firstly the extra 2V on the 5V regulator will increase it's dissipation so it will get much hotter. Also most 10V regulators have a voltage drop of 2V or more. At present the DigiLite board will work happily down to 11V but increasing U2 to a 10V regulator could mean problems with a supply of less than about 12.5V which might not be so good for portable use. A three pin LDO 10V regulator is the answer but is hard to find in an M version. The LM2940T-10.0 is probably suitable. Also the Ultram oscillator will require a separate regulator as it will run happily on 8V but 10V is a little over the top.
Just finished a 24 hour soak test of my DigiLite board with no obvious problems except for a vast video file created by GBPVR
Pity there are no DATV stations near Dover and it's too cold to go portable!
Dick G4BBH
Re: power from digilite on 23cm
Agreed Dick re the portable use but I reckoned that if I was to go portable, with the existing set up that needs a PC for the MPEG encoding, I was going to need mains whether it be from an inverter or a generator. Consequently I would probably do what I did for my G8SUY 23cm FM ATV Tx where I built in a 13.8v 2A SM PSU.
When I first tried the MSA-0886 I wired its bias resistor selected to give 40mA direct to the 12v supply via a choke and some decoupling (not enough initially that caused increased noise floor). That's OK off a stabilized bench supply but no good for battery operation. Now that I've put the 10v regs in I've put a couple of Si power diodes in series with the supply to the Ultram board. Gives reverse polarity protection as well.
I think regrowth is going to be a major problem, especially for those without a SA. It's bad enough trying to get a milliwatt out from a MMIC! I'll be very interested to see what is proposed for the DigiLite amplifier mentioned on the website.
Clive
When I first tried the MSA-0886 I wired its bias resistor selected to give 40mA direct to the 12v supply via a choke and some decoupling (not enough initially that caused increased noise floor). That's OK off a stabilized bench supply but no good for battery operation. Now that I've put the 10v regs in I've put a couple of Si power diodes in series with the supply to the Ultram board. Gives reverse polarity protection as well.
I think regrowth is going to be a major problem, especially for those without a SA. It's bad enough trying to get a milliwatt out from a MMIC! I'll be very interested to see what is proposed for the DigiLite amplifier mentioned on the website.
Clive
Re: power from digilite on 23cm
I have a Toshiba laptop with Windows 7 and a WinTV USB. Portable operation is possible. I had hoped to use the netbook with Windows XP but have failed to get the WinTV USB working on it. The driver fails at every boot and during use. All my other gear has 12V capability (JVC 10" monitors, waveform monitor, TBC etc.).
I tried adding a small buffer amp using a VNA25 and managed to get a reasonable looking output with little regrowth. It drives my RA18H1213G to 1W but not really good enough to put 'on the air'. I will see what happens when I put in the 10V regulator and MSA0886. I only have a couple of these MMICs so will have to be careful
This morning I did the 68pF mod to the Ultram and spurii at +/- 150MHz have gone and also added some other capacitors as discussed on this thread. My noise floor is now over 55dB down from 0 to 1.6GHz and close in.
Dick G4BBH
I tried adding a small buffer amp using a VNA25 and managed to get a reasonable looking output with little regrowth. It drives my RA18H1213G to 1W but not really good enough to put 'on the air'. I will see what happens when I put in the 10V regulator and MSA0886. I only have a couple of these MMICs so will have to be careful
This morning I did the 68pF mod to the Ultram and spurii at +/- 150MHz have gone and also added some other capacitors as discussed on this thread. My noise floor is now over 55dB down from 0 to 1.6GHz and close in.
Dick G4BBH