Speech Compressor - HomeBrew Forum at eHam.net

$\'\'$

$\'\'$

$\'\'$

eHam.net Forum : HomeBrew : Speech Compressor

Forum Help

1-10 of 10 messages

Page 1 of 1

Speech Compressor	Reply
by KJ4KET on June 8, 2009	Mail this to a friend!
Hi: Does anyone know where I could find a simple circuit diagram for a speech compressor with a high impedance input and output? I want to use it with a Heathkit HW-101 and a high impedance microphone. I have seen one or two simple circuits on the internet, but they say nothing about input and output impedance, and I'm not smart enough to tell just by looking at the circuit. David M KJ4KEt

RE: Speech Compressor	Reply
by WB2WIK on June 8, 2009	Mail this to a friend!
It doesn't have to be "high impedance input and output." As a matter of fact, you'd be much better off using a low impedance dynamic microphone with an external speech compressor. A low impedance (output) compressor can drive the high-Z input of the HW-101 perfectly, especially if the output voltage is adjustable. If you find the compressor is overdriving the rig badly, you can always add a series resistor (like 50k) to step the output voltage down slightly for the high-Z input of the Heathkit. The problem occurs in the "other" direction: A high-Z source cannot drive a low-Z load (usually). WB2WIK/6

RE: Speech Compressor	Reply
by G3RZP on June 9, 2009	Mail this to a friend!
A lot depends on what you want the speech compressor for. There are two basic types - the straight compressor, sometimes called a VOGAD (Voice Operated Gain Adjusting Device) and a processor. The VOGAD merely adjusts the drive level to the tx with differing voice levels: popular in some military applications because the guy can whisper but still modulate the radio. Not generally a lot of use in ham radio, it has a long time constant gain control. The other form of compression, the processor, reduces the peak to average ratio of the speech, and so you can run a higher average RF power output for the same PEP. Either type of processing taken to extremes sounds awful: you stop talking and the wife on the 'phone in the other room, the kids playing pop music in the bedroom and the fan on the amplifier all modulate up to full PEP output! Normally about 6 to 10 dB of gain adjustment is adequate. Processors work by producing an SSB signal (can be by the phasing method) clipping the peaks, and then demodulating back to audio and filtering before feeding the tx. Clipping audio and feeding it into an SSB tx doesn't provide the performance. This type of processing can be as effective as small amplifier.

RE: Speech Compressor	Reply
by AA4PB on June 9, 2009	Mail this to a friend!
Also be aware that because the compressor increases the average power output you may need to be concerned about the radio and power supply's ability to handle that power.

RE: Speech Compressor	Reply
by N2EY on June 9, 2009	Mail this to a friend!
I would not recommend using a speech compressor with a stock HW-101. You'll lose more than you gain. The HW-101 is a good rig for its vintage and price but you can't it to work miracles. 73 de Jim, N2EY

RE: Speech Compressor	Reply
by KJ4KET on June 11, 2009	Mail this to a friend!
I was figuring that CW has a duty cycle of 50% or so whereas SSB is lower, maybe 30% or so. My assumption was that some compression or processing could bring up average power to the 50% without endangering the rig, as it would only be working as hard as it would for CW. David

RE: Speech Compressor	Reply
by WB2WIK on June 11, 2009	Mail this to a friend!
D/C for CW is <50% almost always. It would be a bit over 50% if you just sent long strings of "dahs" without any breaks, but there's no reason to do that! Sending standard text with breaks between words, slightly longer breaks between sentences or thoughts, small pauses to think, etc, results in about 30% D/C. SSB without any compression or processing can be quite a bit lower! Most of my rigs measure 100W PEP on a PEP meter while also measuring about 15W peak on an average reading meter (at the same time). Using high level processing can kick the "average" up to 50W or so, while still maintaining 100W PEP. But the rigs are built for it, and the processors are internal, part of the original equipment design. I don't know if an HW-101 would be overstressed by reasonable compression, but I kind of doubt it. Aftermarket speech compressors have been around longer than the HW-101 has! I had a Waters external speech processor back in the sixties, used with a Collins KWM-2 (also a pair of 6146s) and it sounded fabulous and certainly didn't harm the Collins. The 516F2 power supply used by Collins might be a bit more robust than the HP-23 used by Heath, I don't recall. WB2WIK/6

RE: Speech Compressor	Reply
by KJ4KET on June 11, 2009	Mail this to a friend!
YWB2WIK, your comments are more or less in line with my thoughts. The previous commenter is right that one would not want to push old equipment too hard. But it would seem to me that I could provide some beneficial compression without pushing things too hard. Until my HO-10 monitor scope fried (subject of another thread), I could see how except for the peaks of my speech, the envelopes dropped off rapidly so the average power is probably similar to what you found, maybe only 15-20% or so. I have found a freeware program online that is supposed to replicate the effects of an RF speech processor, and I may play with that for a while to see what results I get, once I can make a cable to connect the computer to the rig and to a PTT switch.

RE: Speech Compressor	Reply
by G3RZP on June 12, 2009	Mail this to a friend!
I think a lot depends on the power supply, and a stock HP23 might be a bit light. Depends on what you're doing too - ragchew in a net with short overs occasionally is very different to contesting as far as heat build up in the PS is concerned. G3RZP

RE: Speech Compressor	Reply
by QRP4U2 on June 22, 2009	Mail this to a friend!
I would use a negative peak limiter and not push the old amp too hard. http://amfone.net/Amforum/index.php?topic=17437.0 Phil - AC0OB

Page 1 of 1

Next Topic: Motor rpm counting circuit?
Previous Topic: magnetic loop plans?

Email Subscription
You are not subscribed to this topic.

Check our help page for help using Forum, or send questions, comments, or suggestions to the Forum Manager.

$\'\'$