-
You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.
Rocketbox info thread !!! Made in the USA with a reasonable pricetag !
- Thread starter WINKDOGG
- Start date
I actually found a "Cub Bear 225" that I have had for a few years with a very similar board layout ! I knew the 500 board looked familiar for some reason . I guess unless you get into high dollar amp builds its pretty much the same as most modern electronics . Clones and rebrands , Im always willing to give someone different a shot . I know I have a lot to learn and there is only one way to do that … Bring it up , talk about it and break it down !
Thanks for keeping it positive guys I really appreciate it ! I actually stopped visiting the forums for a few years (all forums) because of internet tough guys and know it alls lol !
I like to hear all of your opinions , facts , stats and knowledge . It restores my faith in forums again
73's and Thanks again Brian 4-2-ZERO In The Garden State !
Hey don't worry Ill take all the stoning's lol Im still ordering that green 500 just to give it a spin !
Thanks Brian
I see nothing wrong with sticking with what works and has worked for decades. I have run/tested probably 25 amps over the years. I most recently had a Palomar 1x4 mosfet amplifier that had the ERF-7530 mosfets. The amp had burn issues on the board by the driver mosfet that had been previously fixed. One or more of the finals got cooked and since the ERF-7530 was discontinued I got rid of it.
All that said if someone can put together a nice mosfet amp that is class AB with a switched SSB dely and priced right I would love to try it. Until then I will stick with what I know works. That video clip on the 9530 data sheet is not real impressive. We will have to wait and see how these Rocketbox's work out.
That 500 looks almost as good as the green lizard 250's that Vern used to build.
But he used real RF transistors.
These things are not designed to be used for RF.
But, someone, is smiling all the way to the bank.....
73
Jeff
But he used real RF transistors.
These things are not designed to be used for RF.
But, someone, is smiling all the way to the bank.....
73
Jeff
I've seen the datasheet many times and this guy is right. It's a joke! If you ignore the value of the coupling cap on the test circuit input, it's an audio amplifier input stage. It's dependent on the driver stages tuned output to be closely coupled to its input since it does not have its own input circuit. As I think about that more, this is probably the schematic from that 75 watt add on heatsink they sell to replace the radios final.
I agree with everything the person making this video has said about the part except I'm confident he's underestimated the interaction between gain and bias with MOSFET's driven at low power levels. If the datasheet is right and the input impedance is really 2.6 ohms with a gate threshold of 4 volts, guess how many watts of drive it takes without any bias just to get the transistor to start conducting? That would be 6.15 watts of wasted drive before the amplifier could produce anything for output.
While the specs are off a little, notice how the first 5 watts of drive only added a watt or two above the input level? Then note how the next 5 watts were able to add 15 more to the input. Proper bias should bring this gain up to 8db or so but that is still very poor for even a bipolar transistor, let alone a modern MOSFET which typically have no problem operating between 17 and 23db today. I understand the desire to be economical but why would you center the entire amp around the cheapest part possible?
For the heck of it, I went to Universal Radios website and pulled the service manuals for several of the latest HF mobile rigs. Rest assured, none use the ERF-9530. There are a wide assortment of fairly priced, real RF transistors still available. If you want bipolar, you have choices from 2SC2904 to 2SC3240. Want to try some new RF 12 volt MOSFET's, look at parts like the RD100HFF1. Mitsubishi has become a leader in this series of low voltage RF parts today.
11 meter mobile operators need to stop copying the amplifier stages of other 11 meter operators and start combining the Push-Pull PA stages used in professional 12 volt communications gear.
Last edited:
Yeah, Brian should have put some gate bias on the thing.
That guy has me puzzled. He has a real engineering background. You would think he would know what happens to an enhancement-mode MOSFET that's being biased only by the drive signal.
Go figure.
73
That guy has me puzzled. He has a real engineering background. You would think he would know what happens to an enhancement-mode MOSFET that's being biased only by the drive signal.
Go figure.
73
Interesting. The one I fixed had a similar failure. The trace between the driver and finals was burned. I overlayed a piece of rg316 and replaced a cap to get it going.
Oh, you ain't shitting that it's high. If I remember correctly, I measured the static gate capacitance at close to 4000pF or 4nF. That's pretty insane. I'm sure with a solid bias and a 9:1 or 16:1 impedance transformer on the gates, you could just brute force the gate to do as you command. Running them in push-pull puts the gates in series and would help with the whole impedance problem a bit, but it won't fix the gain issue.
The internal lead inductance on pins 1 and 3, in this package, the TO247 package, and somewhat lesser in the TO220 package, is pretty high for RF use, and is quite a significant contributor to the gain reduction as the frequency increases. It's not a big issue at the gate terminal, but at the source, it's basically internal negative feedback. Some semi fabs combatted this by using the center pin and the tab as the source connection to greatly reduce this effect, but that isn't possible when Palomar is most likely just relabeling a pre-existing part. I mean, there's a reason that very few legit RF parts come in this, or the TO247 package, it's just not as convenient as it seems at first glance. The ones that I know of off hand are the MRF300xN (50V), and the ARF446G/447G (250V/300V) devices. Both are higher voltage/lesser current devices and both use the center pin and tab as the source, and both have 2 versions with the gate/drain swapped up allow easy push-pull layouts.
And yes, that guy in the video should know full well that these need to be biased to get their full potential, however small that may be.
73s all.
The test is invalid. He knows it was tested with bias and admits to not testing with it.
I said several months ago that Palomar was not a semiconductor manufacturer and I was told that they did. Simple fact companies will rebrand items.
Palomar does not make semiconductors, if they did the Panasonic 2SC2879 would be plentiful. They could buy the discontinued product design and manufacture it under the Palomar brand. I know if I had a facility that could make those transistors I would.
good deal. i think the only valid test is with a real rocket box on a bench.
all this technical stuff is way over me.
You can put a 4 cylinder engine under the hood of any car you want. As long as you know it's a 4 cylinder and drive it like one, it should last. Pull too many RPM's just once, and do not expect it to.
In my mind, the real test for these amplifiers is going to be SSB performance. Even if you only use it on AM, testing it on SSB would reveal any instability or bias issues much faster. The varying drive level of SSB often places unstable amplifiers under the ideal conditions to oscillate. If it doesn't, it would earn a few points of credit from me. You may notice, none are using negative feedback loops either.
In my mind, the real test for these amplifiers is going to be SSB performance. Even if you only use it on AM, testing it on SSB would reveal any instability or bias issues much faster. The varying drive level of SSB often places unstable amplifiers under the ideal conditions to oscillate. If it doesn't, it would earn a few points of credit from me. You may notice, none are using negative feedback loops either.
Last edited:
as much as you know about boxes I'm betting you could look at the inside picture and already know
- No one is chatting at the moment.